Citation
Effect of Cognitive Problem-Solving Style, Internet Usage, and Level of Interactivity on Attitudes toward and Recall of Web-Based Information

Material Information

Title:
Effect of Cognitive Problem-Solving Style, Internet Usage, and Level of Interactivity on Attitudes toward and Recall of Web-Based Information
Creator:
RHOADES, EMILY B. ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Agriculture ( jstor )
Audiences ( jstor )
College students ( jstor )
Gratification ( jstor )
Information recall ( jstor )
Internet ( jstor )
News content ( jstor )
Psychological attitudes ( jstor )
Social interaction ( jstor )
Websites ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Emily B. Rhoades. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2006
Resource Identifier:
649815459 ( OCLC )

Downloads

This item has the following downloads:

rhoades_e ( .pdf )

rhoades_e_Page_016.txt

rhoades_e_Page_018.txt

rhoades_e_Page_061.txt

rhoades_e_Page_156.txt

rhoades_e_Page_051.txt

rhoades_e_Page_100.txt

rhoades_e_Page_131.txt

rhoades_e_Page_008.txt

rhoades_e_Page_004.txt

rhoades_e_Page_023.txt

rhoades_e_Page_118.txt

rhoades_e_pdf.txt

rhoades_e_Page_029.txt

rhoades_e_Page_110.txt

rhoades_e_Page_072.txt

rhoades_e_Page_019.txt

rhoades_e_Page_137.txt

rhoades_e_Page_138.txt

rhoades_e_Page_031.txt

rhoades_e_Page_055.txt

rhoades_e_Page_012.txt

rhoades_e_Page_003.txt

rhoades_e_Page_087.txt

rhoades_e_Page_104.txt

rhoades_e_Page_115.txt

rhoades_e_Page_119.txt

rhoades_e_Page_088.txt

rhoades_e_Page_158.txt

rhoades_e_Page_098.txt

rhoades_e_Page_022.txt

rhoades_e_Page_041.txt

rhoades_e_Page_122.txt

rhoades_e_Page_038.txt

rhoades_e_Page_017.txt

rhoades_e_Page_067.txt

rhoades_e_Page_145.txt

rhoades_e_Page_005.txt

rhoades_e_Page_006.txt

rhoades_e_Page_078.txt

rhoades_e_Page_149.txt

rhoades_e_Page_040.txt

rhoades_e_Page_099.txt

rhoades_e_Page_160.txt

rhoades_e_Page_033.txt

rhoades_e_Page_021.txt

rhoades_e_Page_153.txt

rhoades_e_Page_155.txt

rhoades_e_Page_058.txt

rhoades_e_Page_009.txt

rhoades_e_Page_010.txt

rhoades_e_Page_077.txt

rhoades_e_Page_136.txt

rhoades_e_Page_111.txt

rhoades_e_Page_133.txt

rhoades_e_Page_026.txt

rhoades_e_Page_103.txt

rhoades_e_Page_143.txt

rhoades_e_Page_062.txt

rhoades_e_Page_102.txt

rhoades_e_Page_037.txt

rhoades_e_Page_046.txt

rhoades_e_Page_085.txt

rhoades_e_Page_059.txt

rhoades_e_Page_123.txt

rhoades_e_Page_007.txt

rhoades_e_Page_096.txt

rhoades_e_Page_015.txt

rhoades_e_Page_093.txt

rhoades_e_Page_056.txt

rhoades_e_Page_014.txt

rhoades_e_Page_079.txt

rhoades_e_Page_152.txt

rhoades_e_Page_034.txt

rhoades_e_Page_097.txt

rhoades_e_Page_042.txt

rhoades_e_Page_050.txt

rhoades_e_Page_002.txt

rhoades_e_Page_086.txt

rhoades_e_Page_091.txt

rhoades_e_Page_013.txt

rhoades_e_Page_065.txt

rhoades_e_Page_130.txt

rhoades_e_Page_108.txt

rhoades_e_Page_054.txt

rhoades_e_Page_106.txt

rhoades_e_Page_134.txt

rhoades_e_Page_128.txt

rhoades_e_Page_101.txt

rhoades_e_Page_154.txt

rhoades_e_Page_161.txt

rhoades_e_Page_011.txt

rhoades_e_Page_129.txt

rhoades_e_Page_107.txt

rhoades_e_Page_120.txt

rhoades_e_Page_116.txt

rhoades_e_Page_142.txt

rhoades_e_Page_089.txt

rhoades_e_Page_045.txt

rhoades_e_Page_157.txt

rhoades_e_Page_132.txt

rhoades_e_Page_035.txt

rhoades_e_Page_135.txt

rhoades_e_Page_090.txt

rhoades_e_Page_053.txt

rhoades_e_Page_148.txt

rhoades_e_Page_082.txt

rhoades_e_Page_081.txt

rhoades_e_Page_064.txt

rhoades_e_Page_074.txt

rhoades_e_Page_094.txt

rhoades_e_Page_125.txt

rhoades_e_Page_024.txt

rhoades_e_Page_147.txt

rhoades_e_Page_144.txt

rhoades_e_Page_084.txt

rhoades_e_Page_139.txt

rhoades_e_Page_151.txt

rhoades_e_Page_075.txt

rhoades_e_Page_150.txt

rhoades_e_Page_039.txt

rhoades_e_Page_036.txt

rhoades_e_Page_028.txt

rhoades_e_Page_027.txt

rhoades_e_Page_127.txt

rhoades_e_Page_140.txt

rhoades_e_Page_068.txt

rhoades_e_Page_025.txt

rhoades_e_Page_113.txt

rhoades_e_Page_117.txt

rhoades_e_Page_124.txt

rhoades_e_Page_057.txt

rhoades_e_Page_048.txt

rhoades_e_Page_126.txt

rhoades_e_Page_043.txt

rhoades_e_Page_070.txt

rhoades_e_Page_073.txt

rhoades_e_Page_047.txt

rhoades_e_Page_076.txt

rhoades_e_Page_105.txt

rhoades_e_Page_109.txt

rhoades_e_Page_066.txt

rhoades_e_Page_032.txt

rhoades_e_Page_044.txt

rhoades_e_Page_069.txt

rhoades_e_Page_030.txt

rhoades_e_Page_121.txt

rhoades_e_Page_020.txt

rhoades_e_Page_080.txt

rhoades_e_Page_112.txt

rhoades_e_Page_052.txt

rhoades_e_Page_049.txt

rhoades_e_Page_071.txt

rhoades_e_Page_159.txt

rhoades_e_Page_060.txt

rhoades_e_Page_146.txt

rhoades_e_Page_095.txt

rhoades_e_Page_092.txt

rhoades_e_Page_114.txt

rhoades_e_Page_001.txt

rhoades_e_Page_063.txt

rhoades_e_Page_083.txt

rhoades_e_Page_141.txt


Full Text












EFFECT OF COGNITIVE PROBLEM-SOLVING STYLE, INTERNET USAGE, AND
LEVEL OF INTERACTIVITY ON ATTITUDES TOWARD AND RECALL OF WEB-
BASED INFORMATION















By

EMILY B. RHOADES


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2006

































Copyright 2006

by

Emily B. Rhoades

































This document is dedicated to my husband Aaron. Thank you.















ACKNOWLEDGMENTS

I would like to thank my committee members for all of their time, effort, and

advice through this process. Their insight has made my research stronger. I thank Mindy

McAdams for her wise advice on the Internet and new technologies that she brought to

this study. I thank Brian Myers for his methodological and statistical help and advice on

surviving the process of a dissertation. I thank Ricky Telg for his friendship, mentorship,

and time in helping me through every stage of this study. Lastly, I thank my advisor and

chair, Tracy Irani, for meeting with me endless times when she had many other things

going on. Her advice and guidance have not only strengthened this study, but have

developed me as a researcher and academic.

I am thankful to my friends and family for all of their support through the ups and

downs of the last few years. I thank my friends back home, Marie and Erin, for reminding

me that life is richer than the things I was experiencing in school. It was nice to know I

could always call for a reality check. I want to thank my 310 cohorts for their advice,

friendship, and encouragement. I could not have done it without Shannon, Courtney, and

Wendy, thanks for putting up with me on the bad days and celebrating with me on the

good days. Thanks to everyone for the trips to get ice cream and the weekend nights at

the Jones'. I am indebted to my Florida family. I look forward to many future research

conferences with everyone.









I want to thank my parents for always being on the other end of the phone after a

bad day. I thank them for encouraging me to stay strong and develop myself as a person.

Their guidance and love has taken me further in life than I ever expected.

Finally, I thank my husband for standing by me through the last few years of

school. I appreciate him moving to Florida and putting our life on hold. His dedication to

me and our future is amazing. His friendship and love has encouraged me through all of

this, and I hope someday I can repay him. Thank you.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ....................................................... ............ .............. ix

L IST O F F IG U R E S .... ...... ................................................ .. .. ..... .............. xii

A B STR A C T .............................................................. ...... ..... ......... xiii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

Introdu action to the Stu dy ............................................. ............................ ...........
Com m unication and A agricultural Science.................................................................2
Extension's Role in Communicating Agriculture.............................. .....................5
Internet as an Inform ation Source........................................... .......................... 8
W ebsite D esign and Structure ................................................... ....... ............... 11
U ses and G ratification s............... ............... ...................................... .....................13
Previous U sage with M edia .................................................. ................... 14
A attitude ................................. ........................... .... ..... ........ 15
Information Recall .............................. ............. .. ....................... 16
P ro b le m S o lv in g ................................................................................................... 16
Purpose of the Study .............................................. ... ............................. 19
K ey T erm s ................................................................2 0
O rg a n iz atio n ............................................................................................................... 2 0

2 L ITRA TU R E R E V IE W ...................................................................... ..................22

Overview ............................................................ 22
Interactivity and Linearity ................................................ .............................. 22
In tera ctiv ity ............. .. ............. ................................................................. 2 3
L in e a rity ................................................................................................. 2 8
A daption/Innovation T heory ........................................................... .....................29
Problem Solving ........................ .................. .................... ........ 29
K irton's Theory ........................ ......... .. ........ ............. ..... 30
Inform ation R ichness............ .... ...................................... .... ....................... 36
U ses and Gratifications Theory ............................................................................ 37
Uses and Gratifications and the Internet .................................. ...............39









Previous Experience and Expectations.....................................................41
Inform action Recall ..................................... .............. ......... 42
A attitude .............................................................................44
C onceptual F ram ew ork ...................................................................... ...................46
R research Q uestions............ ................................................................ .......... .... 47

3 M E T H O D O L O G Y ............................................................................ ................... 48

O v e rv iew .........................................................................................4 8
Hypotheses............................................... ........ .................. .......... 48
For Subjects Who Receive the Interactive Site: ......................................49
For Subjects Who Receive the Non-Interactive Site: ....................................49
R research D esig n ..................................................... ................ 4 9
S u objects ...................................... ................................................... 5 2
P ilo t S tu d y ................................................................5 3
P ro c e d u re .......................................................................................................5 4
Instrumentation ............... ......... .......................56
Independent Variables ......... .............. ................. ............... 57
T re a tm e n t ............................................................................................... 5 7
Problem -solving style........................................................ ............... 59
Intern et U sag e ............................................................60
D dependent V ariables .................................................... ........ ..... ............62
Information Recall.................. ............ ..... ... .... ...............62
Attitude ................ ......... .................. 63
D ata A nalysis................................................... 65

4 R E S U L T S .............................................................................6 7

Demographics ...... .................... ... .................68
Media Selection and Internet Usage ................................... ........ 69
E xten sion U sage .............................................................7 1
M essag e R elev an ce ....................................................................................... 7 1
Manipulation Checks ................ ....... .......... ........ 71
P problem -solving Inventory ................................................................................... 72
Internet Usage Constructs ................. ...................74
Attitude Constructs ........................................................................ ......... .......... ........ 75
Inform action R call ...............................................................78
H ypotheses Tests ....................... ........... ................... 79
For Subjects who Received the Interactive Site ................................................83
For Subjects Who Received the Non-Interactive Site .........................................85

5 D ISC U S SIO N ....................................................................... ..................................... 88

O overview .......... .. .......................................................... ...................................... 88
K ey F in d in g s ......................................................................................................... 9 0
Im plications of the Study .......................................................... ............... 94
P problem -Solving Style............. ................................ ...... ........ ............94









Internet Usage .............................. ...... .... ...... ... .... 98
Limitations ........................... ......... ..........100
Recommendations for Theory and Practice..................................... ............... 103
Recom m endations for Practitioners ....................................... ............... 103
Future Research .................... ................... ........ .. ................ 109
C o n c lu sio n s.......................................................................................................... 1 14

APPENDIX

A IN ST R U M E N T S .......................................................................... ....................... 116

B EXPERIM ENTAL CONDITION ........................................ ......................... 125

L IST O F R E FE R E N C E S ......................................................................... ................... 13 1

BIOGRAPHICAL SKETCH ............................................................. ..................146
















LIST OF TABLES


Table pge

3-1. Independent Sample T-test for Significant Differences Between Courses and
Version of the Site Based on Age or Gender ........................................................53

3-2. Means Table for Website Interactivity Identification............... ..........................54

3-3. Independent Sample T-test for Significant Differences Between Less and
Interactive V version of the sites. ........................................ .......................... 54

3-4. Exam ple of K A I Instrum ent Item s ..................................... ......................... ......... 60

3-5. Example of Internet Usage Items. ............. ..................................... 61

3-6. Example of Scale Used to Measure Attitude toward the Treatment or Control
Version of the site to Which Subjects were Exposed .......................................64

3-7. Example of Scale Used to Measure Attitude toward the Internet in General.............64

3-8. Example of Index Used to Measure the Importance of the Internet........................65

4-1. Number of Respondents by Age, Gender, and Class Rank.................... ........ 68

4-2. Number of Respondents by College (n=252)................................ ...............69

4.3. How Participants Access the Internet at Home and at Campus............... ...............70

4-4. Level of General Attitude Toward the Internet ........................................ ...............70

4-5. Mean Extension Experience of Study Participants...............................................71

4-6. M eans Table for Site Interactivity Identification ....................................... .......... 72

4-7. Univariate Analysis of Variance for Significant Differences between Non-
Interactive and Interactive Version of the Sites.....................................................72

4-8. Means Table for Problem-solving Style Based on the KAI ......................................73

4-9. Means Table for the Effect of Gender on Problem-solving Style Based on the
K A I. ................................................................................ 74









4-10. Inter-item Consistency Statistics for the Internet Usage Construct (N=247) ..........74

4-10 Continued. Inter-item Consistency Statistics for the Internet Usage Construct
(N =247) ................................... ........................... ..... ..... ........ 75

4-11. M eans Table for Internet U sage. ........................................................................... 75

4-12. Inter-item Consistency Statistics for the Attitude Toward the Internet in General
(N =249) ................................... ........................... ..... ..... ........ 76

4-13. Inter-item Consistency Statistics for the Attitude Toward the Treatment or
Control Version of the Site to Which They were Exposed (N=237)...................77

4-14. Inter-item Consistency Statistics for the Importance of the Internet in Subjects'
L iv e s (N = 2 3 7 ) ................................................. ................ 7 8

4-15. Descriptive Report for Unaided Recall (N=255).................................................79

4-16. Means for Attitude Toward Treatment/Control split by Low/High Level of
Internet Usage, Adaptive/Innovative Problem-solving Style, and Experimental
Condition Presented (W ith Cell Sizes) ...................................... ............... 79

4-17. Means for Information Recall split by Low/High Level of Internet Usage,
Adaptive/Innovative Problem-solving Style, and Experimental Condition
Presented (W ith Cell Sizes) ...........................................................................80

4-18. MANOVA Results for Problem-solving Style, Site Interactivity, and Level of
Internet Usage on Attitude and Information Recall (N=229)..............................81

4-19. Means for Level of Problem-solving Style and Site Interactivity on Information
R call O v erall .......................................................................82

4-20. Means for Problem-solving Style and Internet Usage on Information Recall
O overall ............................................................... .... ..... ......... 82

4-21. Means for Site Interactivity and Internet Usage on Information Recall ................. 82

4-22 Means for Problem-solving Style on Information Recall............... .... ........... 83

4-23. ANOVA Results for Those Viewing the Interactive Version, Problem-solving
Style, and Internet Usage on Attitude Toward and Information-Driven
E extension W ebsite (N = 109)....................................................................... ...... 83

4-24. ANOVA Results for Those Viewing the Interactive Version, Problem-solving
Style, and Internet Usage on Information Recall (N=115)............................... 84

4-25. Means for Problem-solving Style and Internet Usage on Information Recall for
Individuals Viewing the Interactive Version of the Site........................................84









4-26. Means for Internet Usage Main Effects on Information Recall for Those
Viewing the Interactive Version of the Site .............. ........................................85

4-27. Means for Problem-solving Style Main Effects on Information Recall for Those
Viewing the Interactive Version of the Site .............. ........................................85

4-28. ANOVA Results for Those Viewing the Non-Interactive Version of the Site,
Problem-solving Style, and Internet Usage on Attitude Toward and
Information-Driven Extension W ebsite (N=121)............................................... 85

4-29. ANOVA Results for Those Viewing the Non-Interactive Version, Problem-
solving Style, and Internet Usage on Information Recall (N=115)......................86

4-30. Means for Problem-solving Style and Internet Usage on Information Recall for
Individuals Viewing the Non-Interactive Version of the Site. ............................87

4-31. Means Problem-solving Main Effects on Information Recall for Those Viewing
the Non-Interactive Version of the Site. ................. .......................87
















LIST OF FIGURES

Figure page

2-1. Conceptual framework for this study. ............................................. ............... 47

3-1. A screen capture of the consent information and instructions sent to participants. ...55

3-2. A screen capture of the non-interactive control page sent to participants ................57

3-3. A screen capture of the interactive version of the site page sent to participants........58















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

EFFECT OF COGNITIVE PROBLEM-SOLVING STYLE, INTERNET USAGE, AND
LEVEL OF INTERACTIVITY ON ATTITUDES TOWARD AND RECALL OF WEB-
BASED INFORMATION

By

Emily B. Rhoades

August 2006

Chair: Tracy Irani
Major Department: Agricultural Education and Communication

This study examined the effects of problem-solving style, level of Website

interactivity, and Internet usage on an individual's attitude toward an information-driven

Extension Website and subjects' recall of the information presented on that site. This

study is based on a conceptual framework relating Kirton's Adaption Innovation Theory

and Uses and Gratifications.

Successful problem-solving is in demand in the area of agriculture. As Extension

services and communicators move to designing information online, it is crucial that this

information be presented in a form that will be usable, valuable, appropriate, and easy to

recall. By understanding how problem-solving styles affect users' perceptions of

Websites, with respect to such attributes as attitude and recall information, Extension,

agricultural communicators, and commodity groups who are utilizing the Internet to









reach audiences will be better able to develop communications processes that match

audience needs in order to inform them, educate them, and effect productive change.

This study shows that problem-solving styles, coupled with an individual's Internet

usage have an affect on information recall. While researchers continue to debate if

interactivity affects attitude and recall of information, these findings show no individual

effects of interactivity on attitude and information recall when presenting information-

driven content to a young adult population. However, it was found that for interactive or

non-interactive versions of an information-driven site, information recall could vary

based on problem-solving style and level of Internet usage.

There are populations such as innovative problem-solvers who retain information

better from the non-interactive versions of online Extension information. The more

adaptive individuals will actually do better with less structure and ambiguity when

working online successfully. It is also noted that for low users of the Internet, the novelty

of interactivity attracts and keeps the interest of users to increase their retention of

information, as supported by the literature. These findings encourage designers of

information-driven sites to take inventory of how they are presenting their information to

specific audiences.














CHAPTER 1
INTRODUCTION

Introduction to the Study

"A cow that was cleared of having mad cow disease last fall by the U.S.
Department of Agriculture was in fact infected with the brain-wasting disease, the
department announced Friday, making it the second confirmed case of the disease
in this country" The Chicago Tribune reported on June 25, 2005 (U.S. Confirms
Mad Cow, 2005).

Communicating information about science has always been an important endeavor,

but perhaps never as crucial as it is today. In an age which the technology to create the

science seems to be in a race to outpace the technology used to communicate it,

communicators must be set to address the changing needs of audiences. Today, the public

is faced with many information choices, presented through a multitude of different

communication channels. One such channel that has emerged as an important tool for

those seeking science information is the Internet (Morris & Ogan, 1996).

Science-based information has the potential to create situations that save lives

(Henroid, Ellis, & Huss, 2004). After completion of formal education, most people will

only be exposed to science through chance encounters with news reporting (Treise &

Weigold, 2002) and occasional informal education. Much of the information presented

through mass media informs audiences about breakthroughs in science, food safety,

medicine, and technology, all topics that have the potential to greatly influence lifestyles.

While science information is usually thought of as being related to biology,

biotechnology, food science, or horticulture, science topics covered by organizations like

Extension also include social sciences and economic research. One such topic covered by









Extension science information includes budgeting and purchasing of large items like

automobiles (University of Iowa, 1998).

The Internet can offer new challenges to those communicating science-based

information to audiences (Henroid, Ellis, & Huss, 2004). Through the power to integrate

other forms of media on the Web, communicators are now able to present complex,

science-based information in new ways, such as interactive video, animations, and

graphics, which create memorable learning experiences for viewers. Finding and

retrieving information is considered to be more convenient by online users who are able

to find information on many topics in a short amount of time (Henroid, Ellis, &Huss,

2004). Along with these new opportunities, however, come many challenges. Research is

just beginning to examine the answers as to how user characteristics relate to the

processing of design and display of information online. Given the importance of

effectively conveying scientific information, researchers must examine the best methods

available to help people successfully discover and interpret the information they find

online.

Communication and Agricultural Science

Today's world is science-driven, and for the benefits of scientific advancements to

be dispersed, publics need to be able to interpret and understand that information

(Shortland & Gregory, 1991). A society's understanding of this information is important

not only for the well-being of its citizens, but also for the continued support of these

endeavors. Educated publics should be able to choose between conflicting reports on

information concerning scientific advancements (Treise & Weigold, 2002). By

effectively communicating science to audiences, favorable attitudes that are created









toward science and science funding will allow for a clearer understanding of the benefits

that science adds to society (Treise & Weigold, 2002).

Scientific information tends to be complex, detailed accounts of new advancements

or findings. Research has shown that science communicators often frame the news by

only reporting on the breakthroughs (Gunter, Kinderlerer, & Beyleveld, 1999). For

example, researchers, looking at the opinions of scientists and journalists, found that both

groups agreed media coverage of biotechnology information was questionable, taking

into account the complexity of the subject (Gunter, Kinderlerer, & Beyleveld, 1999).

Poor reporting of this information has been a concern of scientists and researchers alike

(Gunter, Kinderlerer, & Beyleveld, 1999; Treise & Weigold, 2002).

As a subset of the scientific community, agricultural science is an important aspect

of science with respect to America's economy and environment (Ruth, Telg, Irani, &

Locke, 2004). Many current scientific issues that have been extensively reported on, such

as mad cow disease, biotechnology, and animal cloning, are all deeply embedded within

agriculture.

Developments in agriculture over the last few years have created many

opportunities, as well as challenges, to researchers and communicators. Agricultural and

science information affects everyone on an everyday basis, whether they are aware of it

or not (Saunders, Akers, Haygood, & Lawyer, 2003; Lundy, Ruth, Telg, & Irani, 2005).

It is vital that this information is perceived accurately by the general public, due to the

significant impact of agriculture on society and public health (Terry & Lawyer, 1995).

For generations, agriculture has been intertwined with greater human society by

serving as a support and underpinning (Pawlick, 2001). However, as important as the









information is, many argue that agriculture science is minimally covered through the

media (Pawlick, 2001). "The changes in agriculture and its impact on the American

economy make the need for communicating agriculture crucial for creating an

agriculturally literate public" (Lundy, Ruth, Telg, & Irani, 2005, p. 6). Frick, Birkenholz,

and Machtmes (1995) also contend that every person should possess a minimum amount

of knowledge and understanding of the scientific industry that provides food for human

survival. The decreases in the farming and ranch populations have made this an even

more vital need, as members of the general public, far removed from the rural setting,

may no longer have accurate perceptions of agriculture (Saunders, Akers, Haygood, &

Lawyer, 2003). Consumers need to be literate on agricultural issues in order to respond

aptly when issues that deal with the intersection of food and science occur. For this to

take place, however, these issues must be communicated effectively.

The agricultural industry, the federal and state-mandated land grant institutions, the

farm press, and academic journals continue to disseminate information on agricultural

issues. Land grant institutions in particular, through their Cooperative Extension Services

and public information specialists, are charged with the task of technology transfer and

disseminating public-value information to their clientele and members of the general

public (Seevers, Graham, Gamon, & Conklin, 1997). Until the advent of the Internet,

Extension communicators were forced to use traditional mass media as their primary

communications channel. However, traditional media gatekeepers, often focusing on

news value as a function of their audience's demographics and occupations, have been

viewed by some as not always highly valuing agricultural news, and information with









respect to determining what is on the public, and consequently, the media agenda

(Cartmell, Dyer, Birkenholz, & Sitton, 2004).

Extension's Role in Communicating Agriculture

As discussed earlier, one such organization charged with communicating science to

audiences is the U.S. Cooperative Extension Service. Extension links education and

research resources of the land-grant university, the United States Department of

Agriculture (USDA), and county administrative units to provide scientific information to

non-formal educational audiences (Seevers et al., 1997).

The Smith-Lever Acts of 1862, 1890, and 1990 formally integrated these resources

to provide instruction and information in agriculture and home economics and related

subjects to those not currently taking courses in the Land Grant colleges. The land grant

College system was established through the Morrill Acts in 1862 and 1890 to provide at

least one college in each state that would have a leading objective to provide learning

opportunities related to agriculture and the mechanic arts to the industrial class (Seevers,

Graham, Gamon, & Conklin, 1997).

Cooperative Extension has changed its focus over the years to include broader

initiatives and issues by serving as problem solvers in issues of the environment and other

social and economic changes going on in communities (Seevers et al., 1997). While

Extension programs were historically seen as rural programs, they now cover almost

every aspect of people's lives. They have also begun moving to reach greater audiences,

including rural and non-rural audiences of all ages through new technological and

communication outlets (Seevers et al., 1997).

Researchers over the last several years have been providing evidence of

Extension's need to embrace Internet technologies to reach audiences (Howell & Harbon,









2004). The broader audiences and community-based needs being addressed by Extension

in the 20th century caused researchers and Extension professionals to look into new and

more cost-effective methods of information dissemination (Bull, Cote, Warner, &

McKinnie, 2004; Wood-Turley & Tucker, 2002). The need to provide these diverse

audiences with timely, pertinent information that allows them to maintain a working

dialogue with these audiences has caused many Extension communicators to develop

communication via the Internet (Siegrist, Labarge, & Prochaska, 1998). Several studies

on Extension have encouraged the continued movement online. Kaslon, Lodl, and Greve

(2005) studied the effectiveness of online leaders training for 4-H volunteers, and found

that the Internet is a good method to reach these audiences with training and continued

education. Lippert, Plank, and Radhakrishna (2000) looked at the effectiveness of

regional Internet Extension in-service training to reach agents. The researchers found that

it was not only successful in knowledge acquisition, but was also seen as just as effective

as face-to-face administration by participants. Dunn, Thomas, Green, and Mick (2006)

found that an interactive online multimedia Extension product can increase knowledge

and influence behaviors on nutrition with high school students. They recommended that

multimedia was a good way for Extension to educate young people on health-related

topics.

In a consumer focus group study focusing on the value of Extension services, Irani,

Ruth, Telg, and Lundy (2005) recommended that Extension adopt the communication

technology used by their target audiences. The researchers also noted that for the

participants they assessed, that technology was the Web. The ability of the Internet to

provide cost efficient information that can reach large audiences has been described as a









valuable tool for Extension and its clientele, especially with younger audiences (Jackson,

Hopper & Clatterbuck, 2004). A study of landowners in 2004 indicated age was a

significant factor as to whether the public wanted information on watershed conservation

issues via the Internet or through written communication. Younger landowners showed a

higher preference for computer-based information (Howell & Harbon, 2004), while the

majority of respondents still preferred traditional written communication. Howell and

Harbon (2004) concluded that while current trends may still prefer traditional

communication methods, Extension must move toward targeting these younger audiences

who will be the landowners of the future.

Bull and colleagues, in 2004, called for Extension to pay specific attention to

underserved audiences who have typically not been original stakeholders in the program.

Young adults ranging in age from 18 to 24 are among one of the groups not traditionally

serviced through Extension programs that are traditionally aimed at youth, pre-college,

and adult homeowners (Seevers, et al, 1997). While research has shown that younger

adults are some of the lowest users of Extension, they are interested in Extension services

such as community development programs (Warner, Christenson, Dillman, & Salant,

1996). Audiences of young adults are also the future users of Extension programs as they

graduate and become involved in communities and purchase homes.

Nationally, Extension has attempted to answer this call to move online by

introducing the e-Xtension initiative, led by the Extension Committee on Organization

and Policy (MSState, 2005). The goal of this program is to implement a national Web-

based information and education network for all Extension clientele (MSState, 2005).

This modern marketplace will facilitate engagement with audiences in new subject areas









in a manner that is accessible and timely. Users of the system will be prompted to provide

information that allows them to receive personalized assistance for the information

contained in the 3,000-plus counties in the U.S., and yet still be connected to their state

and local Extension organizations (MSState, 2005).

By focusing on answering users' questions and problems, plans are for e-Xtension

to provide information in various interactive formats, including frequently asked

questions, fact sheets, chat sessions, discussion boards, streaming video, Web-based

conferencing, and educational modules (e-Xtension, 2005). By providing convenient,

quick-access to information, the goal is that users will be able to solve problems and find

information to improve their daily lives, thus supporting the mission of Extension. The

foundation for this initiative was set in place and in September 2005, communities of

practice, or topic areas that will be focused upon, were developed. These included

parenting, horticulture, disaster education, financial security, local economics and

entrepreneurship, wildlife management, fire ants, and equine resources (e-Xtension,

2005).

Internet as an Information Source

One reason organizations like Extension are moving online is because of the

dramatic increase in Internet usage. Nielsen/NetRatings reported that 135.82 million

Americans were active Internet users in 2004 (ClickZ, 2005). As of February 2003,

Americans spent an average of 25.5 hours per month using the Internet (CyberAtlas,

2003).

Recent studies have shown that one major use of the Internet by these populations

has been for news information. The Web has been found to be the third-most important

source of news following radio and newspapers (Chan & Leung, 2005). It was noted in









2003 that 40% of adults in the United States use the Web to get news, weather, and sports

information (Lieb, 2005). A survey done by the Pew Research Center found that one-

third (and almost half of those under 30) of respondents now receive news information

online at least once a week (Bogart, 2000). Conway (2001) reported that more than four

out of every 10 respondents to his study were using computers to find out what was

happening in the world. In a survey of 400 Midwest university students, it was found that

47.8% use the Internet frequently for reference or research materials (Bressers & Bergen,

2000). Stempel, Hargrove, and Bernt (2000) found a decline in the use of local and

network television and newspapers while there was a huge gain in Internet use by the

general public. Chan and Leung (2005), on the other hand, found that heavy users of

newspapers and radio tended to spend a longer amount of time online reading news than

light users.

Communicators are attempting to reach their changing audiences by offering online

news and information. Garrison (2001) found, that as of 1999, almost 90 % of U.S. daily

newspapers were actively using new online technologies to reach new markets. It can be

assumed this number only continues to grow as more users are logging on.

In a national study done by the Pew Internet & American Life Project, 80% of all

Americans said they would expect to find information online about health and news

(Horrigan & Rainie, 2002). One in five Americans revealed that they rely heavily on the

Internet to find information. Of those who go to the Internet for government and health

information, the majority was white females with children under the age of 18 (Horrigan

& Rainie, 2002). According to a recent survey of Internet users, 49% are college

graduates, 38% have family incomes over $75,000, and their main transaction online is to









gather information or for entertainment (Fallows, 2004). Men are more likely than

women to use the Internet for information, while women use it more to communicate. In

2005 it was noted that 68% of males and 66% of females were Web users, and 80% of

males and 86% of females ages 18-19 were using the Internet (Burns, 2005).

College students represent the largest population of Internet users (Eastin, 2001)

making them an important subset to study. An overwhelming 96% of all 18-29 year-old

users find the Internet a good way to get information, compared to 91% of all older users

(Fallows, 2004). And with more than 15 million college students in the United States who

represent a $9.2 billion market for consumer goods (Ness, Gorton, & Kuznesof, 2002)

this audience is one to pay close attention to. Fallows (2004) found in a recent survey of

Internet users that 49% are college graduates, making future graduates an important

audience segment to study.

When the Internet began growing as a communication outlet, Morris and Ogan

(1996) called for scholars to rethink definitions and categories of communication and

mass media in terms of the new technology. Webster and Lin (2002) also found that the

Internet is a viable communication outlet that should be looked at by researchers.

Scholars have answered this call with research in this area; however, many questions

remain unanswered.

Dibean and Garrison (2001, p. 88) concluded, "development of the technology of

the Internet and the Web itself may become the most significant change in world

communication in a half-century or longer." Based on the growing influence of this new

medium, researchers and communicators need to understand how users best process









information presented online. The Internet has created the opportunity for new methods

of news delivery by combining components of print and broadcast media (Berry, 2001).

Although use of the Internet has been studied in the context of news and

information dissemination in general, limited research has been done in the area of

agricultural communications, in particular, in terms of providing information online that

is preferred and is recalled by audiences. Davis and colleagues (2005) found that adults

studied recalled agricultural communication information better when presented to them in

a print, video, or radio news release over electronic text. Based on a thorough review of

the literature, however, few studies have assessed how the level of interactivity of online,

Web-based Extension communication efforts and one's psychometric traits, such as

problem-solving style, affects the information recall and attitudes of users.

Website Design and Structure

Many researchers have begun to study the Internet from a visual communication, or

graphical and structural, perspective in order to analyze how Websites are using design to

reach general audiences, as well as how certain components on these sites aid in recall of

information and perceived preference (Cho, 2003; Bogart, 2000; Diao & Sundar, 2004;

Lang, Borse, Wise, & David, 2002). Abraham (2001) argued that online communication,

by its very nature, is a presentation that is driven by visuals and visual communicators.

Esrock and Leichty (1999) call for communicators to think of their users and to

develop sites that are not only efficient in terms of technology, but also visually pleasing.

In a medium that allows for displaying graphics and multimedia, it is easy to provide

information on pages that are pleasing to view and easy to navigate (Henkia, 1990). Few

people want to dig through confusing pages of information, and designing a site that is









easily navigated can complement communications by keeping viewers coming back

(Henkia, 1990).

Users are looking for simplicity and usability as they enter sites (Nielsen, 2000).

Swenson, Constantinides, and Gurak (2002) described a need to use logical design

choices and define the audience members of the site in order to better reach them.

Websites are a visual media, in which factors such as layout, design, and graphics can

either add credibility to an organization and aid in information intake, or hinder

individuals' ability to process information (Amant, 2005).

Recent studies in the area of Internet media have focused on the effect of non-linear

based information (Lowrey, 2002; Dimitrova, Connolly-Ahern, Williams, Kaid, & Reid,

2003; Tremayne, 2004). In one such study, Lowrey (2002) found that non-linearity did

not affect perceived credibility or knowledge acquisition. Tewksbury and Althaus (2000)

compared the non-linear reality of online news to print editions of the same newspaper.

Findings showed that online readers were less likely to recall events of national,

international, and political importance than those reading traditional print-based

publications.

In agriculture, the research has been less extensive in this area; the look and feel of

sites hosted by communication organizations has been under-researched (Williams &

Woods, 2002). In a research synthesis of the Journal of Applied Communications from

1992-2001, Williams and Woods (2002) noted that a large portion of published research

analyzed the readership trends of agricultural communication outlets, but has ignored

their design or Web presence. Researchers have targeted agricultural communicators

through practitioner-oriented articles in the Journal of Applied Communications, the









leading journal in this applied field, written to help them design effective Websites for

their audiences, but little research has been presented on designing effective sites using

features such as interactivity (Emery, 1999; Kelleher, Henley, Gennarelli, & Hon, 1997;

Melgares, 2005).

Uses and Gratifications

Researchers have described communication behavior as being goal-directed,

purposive, and motivated (Rubin, 1994). The Web has been described by several scholars

as being a medium that requires active users (Bouwman & Wijngaert, 2002; Kaye &

Johnson, 2002; Papacharissi & Rubin, 2000). Users of the Web are challenged with

finding the information that brought them to that site.

Uses and gratifications, as a theoretical perspective, describes users' psychological

and social environmental needs, their needs and motivations to communicate, the media

they choose, their attitudes toward that media, the alternatives to that media, their

communication behavior, and their outcomes are all important elements in the

communication process (Rubin, 1994). By initiating the selection of and use of a specific

media vehicle, users are actively seeking out information in order to fulfill a need (Rubin,

1994). The most salient use of the Web seen by researchers has been the information-

seeking function (Papacharissi & Rubin, 2000). This theory can play a role in assessing

Web use as a function of sociability, prurience, curiosity, and information-seeking

(Ruggiero, 2000).

Uses and gratifications discusses the cognitive processes that take place between

the complexities of needs felt by individuals, such as solving problems or making

decisions and how users gratify those needs through media (Blumler, 1979). Graber

(1984) argues that those who are drawn to media information to receive gratification are









more likely to be able to learn that information. In order to discover what aids people in

recall when using a specific media, it is imperative that researchers keep in mind the

motives that bring users to that media and how they perceive it. Recall has shown to

differ as a result of media consumed (Davis et al, 2005; Tewksbury & Althaus, 2000;

Eveland & Dunwoody, 2001), and by understanding what brings users to media, it can

aid in encouraging users to go to the right kind of media that will help them recall

information.

Uses and gratifications scholars have examined many motives for using the

Internet, and found in general that the Internet tended to satisfy entertainment,

information, and interaction needs (Papacharissi &Rubin, 2000; Ko, 2002). Kaye and

Johnson (1998) found that 40% of respondents used the Web primarily for information

and education research. Kaye and Johnson (2002) concluded that respondents were active

users who sought out specific information via searches and interacted with others through

chat rooms and listserves. Previous Uses and Gratifications studies found, political

attitudes were strongly linked to measures of information seeking and surveillance (Kaye

& Johnson, 2002).

Thus, through the advent of the Internet, Uses and Gratifications researchers have

begun retesting items found to be salient with respect to other media (Ruggiero, 2000).

Ruggiero argued that the advent of the Internet would only increase the theoretical

potency of Uses and Gratifications by allowing researchers to explore the theoretical

linkages with respect to this new communications medium.

Previous Usage with Media

According to the Uses and Gratifications theory, previous experience and

gratifications met can give users a respective image of that medium and what they can









expect from it (Katz, Blumler, & Gurevitch, 1974). Peled and Katz (1974) found in a

study of media during wartime and crisis that people came to a specific medium with

expectations of what that medium will be and what it will gratify for them. While a large

percentage of usage and gratifications research has explained previous usage in the

context of traditional media channels, much new research is being conducted with the

Internet (Ruggiero, 2000; Baran & Davis, 2003). Many recent studies have utilized Uses

and Gratifications as part of the theoretical framework when studying the Internet

because of the interactivity, demassification, and asynchronicity it allows that other

media outlets do not (Baran & Davis, 2003). With 135-plus million users actively using

the Internet in 2004 (ClickZ.com, 2005), it could be the case that they are all coming to

the medium with preconceived ideas of the qualities of that media source.

Attitude

Within the Uses and Gratifications paradigm, attitudes are formed based on the

experience and the gratifications met as a function of choosing a particular medium to

serve a specific need or information-seeking function. Attitude has been shown to be an

important predictor of usage and implementation of technology and continued use

(Rodgers & Chen, 2002). Attitude research has been done extensively in the area of

advertising and attitude toward advertisements and their effectiveness (Chen, 1999;

Rodgers & Chen, 2002; Sinclair & Irani, 2005; MacKenzie & Lutz, 1989), and

researchers have begun employing these same tasks to look at Internet sites and

advertising (Rodgers & Chen, 2002; Chen, 1999). Rodgers and Chen (2002) reported

adoption of the Internet by advertising agencies is affected by poor attitudes toward and

lack of experience with Internet advertising. Chen (1999) developed a scale to provide

researchers the ability to measure the attitudes of users of Websites to help indicate the









value of such sites. For the purposes of this study attitude was defined as: "A

psychological tendency that is expressed by evaluating a particular entity with some

degree of favor or disfavor" (Eagly & Chaiken, 1993, p. 1).

Information Recall

As discussed earlier, recall can be increased as a function of the enhanced

gratification from a medium. Several studies have looked at the Internet and information

recall. Lowery (2002) reported that linearity of sites had an effect on degree of perceived

control over media, but it did not affect the degree of perceived credibility or recall of

knowledge. D'Haenens, Jankowski, and Heuvelman (2004) reiterated this, saying that

news category, gender, and interest played more of a role in recall than whether the

information was given via online or in print. For the purpose of this study information

recall is defined as the learning process and free recall of information, not just aided

identification.

Problem Solving

Problem solving has been described as a survival skill in today's technological

world (Wu, Custer, & Dyrenfurth, 1996). Kirton (2003) described problem-solving as the

means by which life survives and manages the constant change presented through one's

environment. Problem solving has also been "defined as the tendency to respond in a

certain way while addressing problems" (Wu, et al, 1996, p. 55). This linear process of

evaluation begins with users recognizing a specific problem, defining it, having the

ability to comprehend and develop it, test hypotheses and gather data about it, revise

those hypotheses and retest, and then form a conclusion on the problem (Hedges, 1991).

Part of the process of problem-solving includes gathering information through channels

such as the Internet. As one begins this problem-solving process, it is important to note









that individuals are limited by the way they are built in terms of intelligence, but they

also have no instincts to help or hinder them in this (Kirton, 2003, p. 33). However,

individuals are intelligent within different styles that allow them to problem solve given

the opportunity or motivation (Kirton, 2003). As they work through this linear process of

problem-solving they are driven by our individual problem-solving style (Kirton, 2003).

Style is something that is unique to individuals as a psychometric quality while process is

the structure in which all individuals go about solving problems.

A great deal of research has been done on the cognitive and decision-making

processes that bring people to specific media. One aspect of this approach has been to

look at the concepts of information richness and equivicality. Information richness theory

claims that individuals choose media they perceive to be most efficient in helping them to

complete a communication task or problem (Kelleher, 2001). People cognitively choose

which medium will help them as they work to solve problems or gratify other needs. A

rich medium can be described as containing more face-to-face type interactions, while

leaner media are seen as being formal and numerical, such as telephones versus fliers or

fact sheets (Trevino, Daft, & Lengel, 1990). Those in a high equivocality condition will

tend to choose a richer Website more often than leaner communications media, such as a

pamphlet or a lean Website (Irani & Kelleher, 1997). People will choose rich media

based on equivocality, which can be assumed to be related to the innate differences in

which people solve problems. Trevino, Daft, and Lengel (1990) further described

equivocality by comparing an equivocal message to a Rorschach ink-blot-multiple users

may read it differently than others depending on their unique backgrounds and

perspectives. This theory relates to many of the current components of interactivity that









are seen on Websites. As a Website has more interactivity, it will be richer and have more

equivicality, as opposed to a leaner site with more ambiguity.

Another way researchers look at the phenomenon of how individuals choose media

to solve problems is through cognitive styles. "Cognitive activity refers to the degree of

mental activation invoked in paying attention to a medium" (Gunter, 2000, p. 164).

Cognitive style has been described by researchers as also referring to "a person's

consistent pattern of processing information and organizing it into a system of thought

which influences behavior" (Foxall & Haskins, 1986, p. 65) when they are working to

solve problems or make decisions. The cognitive framework allows us to explore the

levels of involvement with media (Gunter, 2000).

Much research has been conducted in this area examining the visual representation

of broadcast news; however, some researchers have begun to use these same methods to

analyze online information (Fox et al., 2004; Lang, Borse, Wise, & David, 2002; Diao &

Saunder, 2004; Gunter, 2000). Gunter (2000) stated that to fully understand the use of

media, one must assess the nature of the exposure in terms of the cognitive effort put into

the processing of the content. Once this is known, one can assess and measure the media

influence on awareness and knowledge gained by the audience (Gunter, 2000).

Researchers have tested specific media outputs to determine the amount of

information recalled and the cognitive effort placed into viewing. Consideration has been

given to how information is cognitively processed and recalled based on the effects of

how that information is presented (Gunter, 2000). Sicilia, Ruiz, and Munuera (2005)

looked at the effects of Internet non-linearity with respect to need for cognition, or an

individual's tendency to engage in and enjoy cognitive activity (Petty & Cacioppo, 1979).









Sicilia and colleagues (2005) found that individuals who were exposed to interactive

Websites processed information more thoroughly than those exposed to non-interactive

sites. For example, people looking at a site that was non-linear and required interaction

would process the information presented more thoroughly than those presented with a

linear site.

One researcher who has focused on problem-solving styles based on cognitive

processing is Michael J. Kirton. Kirton (2003) states that people can be placed on a

continuous scale, with individuals who are adaptive in their problem-solving style at one

end and those who are innovative in their problem-solving style at the other end. This

theory defines and measures the thinking style that influences one's decision-making

process (Kirton, 1999). According to Kirton, the adaptor will solve problems within his

or her existing perceptual frame of reference, while innovators will change those

frameworks and do things differently as they seek solutions outside of the context of the

given problem (Goldsmith, 1984; Foxall & Haskins, 1986). Researchers measure the

adaptive-innovative dimension of cognitive style with the Kirton Adaption-Innovation

Inventory (KAI) (Foxall & Bhate, 1993).

Purpose of the Study

Problem solving has been discussed in various contexts; however, limited research

has been conducted on how problem-solving style aids in the decision-making process

with respect to processing media that may provide information, especially information

that can be used to solve problems from the standpoint of recall of information and

attitude toward the value and appropriateness of the information and media utilized.

The study postulates problem-solving style, as conceptualized by Kirton's KAI

inventory, will influence the way information seekers go about fulfilling their information









needs based on previous knowledge/use of media, and in turn will influence their

attitudes toward the type of media that users prefer and gain knowledge from.

One such area where problem-solving is in demand is the area of agriculture. As

Extension services and communicators move to designing information online to reach

audiences and inform them about topics that will help solve problems, it is important that

this information be presented in a form that will be usable, valuable, appropriate, and

easy to recall.

By understanding how problem-solving styles affect users' perceptions of

Websites, with respect to such attributes as attitude and information recall, Extension,

agricultural communicators, and commodity groups who are utilizing the Internet to

reach audiences will be better able to develop communications processes that match

audience needs in order to inform them, educate them, and affect productive change. If

users vary due to their problem-solving style, it may be the case that communicators need

to provide information that appeal specifically to these diverse styles.

Key Terms

Information-Driven Website: "The objective of information-driven Websites is to
provide the user with desired information. The objective of these sites is to guide
the user to the desired content pages. Navigation pages support the user in his
search" (Stolz, Viermetz, Skubacz, & Neuneier, 2005, p. 1).

Interactivity: "A process involving users, media, and messages, with an emphasis
on how messages relate to one another" (Sundar, Kalyanaraman,, & Brown, 2003,
p. 34).

Organization

Chapter 1 has introduced the problem to be examined, as well as the purpose of the

study. Chapter 2 will discuss in detail the relevant literature and theoretical framework to

be used in the study. These will include the Adoption Innovation model, Uses and






21


Gratifications, Website interactivity and usage, attitude, perceptions of the Internet, and

information recall. Chapter 3 will outline the research design and methods of the study,

including hypotheses, independent and dependent variables, description of participants,

instruments and reliability of scales, procedures, and statistical analysis used. Chapter 4

gives the results of data analysis performed to test stated hypotheses. Chapter 5 will

describe the limitations of the study, the results, and provide conclusions and

recommendations.














CHAPTER 2
LITERATURE REVIEW

Overview

The purpose of this investigation was to examine the effects of individuals'

problem-solving style, level of Website interactivity, and level of Internet usage on

subjects' attitude and amount of information recall with respect to an information-driven

Extension Website. By understanding how problem-solving styles affect users'

perceptions of a Website with respect to such attributes as attitude and information recall,

Extension, agricultural communicators, and commodity groups who are utilizing the

Internet to reach audiences will be better able to utilize communications processes to

inform audiences, educate them, and affect productive change.

The following literature review explores the various components of cognitive

problem-solving style as it relates to the study, as well as what drives individuals to

choose specific media in certain situations, what design factors can be manipulated in an

online environment by the organizations posting the information, and what influences an

individual's attitudes and level of information recall.

Interactivity and Linearity

A significant amount of research in the advertising, communication, and marketing

literature has focused on Website aesthetics, usability, and design. Within these domains,

many researchers have addressed the question as to how to make Websites more

appealing to audiences. Resnick and Montania (2003) used the effects of semiotics, the

study of signs and visuals, in Web design features to explore the expectations of









performance criteria in a purchase situation. The authors found that some design features

have a strong effect on expectations of consumers. Cho (2003) found in a study of online

banner advertisements that peripheral cues such as advertisement size and animation had

an effect on those likely to click-through when they had high involvement with a product,

while Thompson and Wassmuth (1999) cautioned that the use of trick banners on sites

might lead to possible negative reactions.

By ensuring good usability, design, and easy navigation through a site,

communicators and Web developers can attend to their audiences. As users feel more

comfortable with a site and are successful in gratifying their needs, they will like that site

more, and they will return to the site again for information (Spool et al, 1999). While a

site may be designed to effectively reach audiences with information, it is still up to the

user as to how they use that information. As Nielson (2000) described, users have been

shown to scan information online as opposed to engaging in deep cognitive processing. It

has also been noted by authors in the field of Website design that no two users will have

the same experience with a site (Krug, 2000).

Interactivity

Interactivity is something that has been defined by various scholars in many

disciplines to mean different things. Heeter (1989) entered the discussion early on

commenting on how mass media has changed with the onslaught of new media like the

Web. He discussed how the idea of mass has changed as those who look at a Website

may not see the same thing as someone else looking at the same site, due to what he

describes as interactivity and hypertext. Heeter (1989) set up several components of what

he determined was a multidimensional concept, the first being complexity of choices

available to the user, which is defined by how many opportunities there are for users to









make decisions and to be in control and active. Along that same line is what he described

as the effort exerted by the user. He also described that interactivity takes into account

information relative to user's needs and the feasibility to have interpersonal

communication. Heeter said that the measuring the users interactions while online also

comes into play. And lastly, the ease of adding information is also to be considered.

Beyond those steps Heeter went on to describe how this new component of media

changes the role of sender and receiver by making them interchangeable.

Downes and McMillan (2000) added to this definition of interactivity. While they

too expressed the importance of the changing role of sender and receiver and the

importance of choice and effort exerted, they went further with this thought. They looked

at interactivity in terms of its impacts, its nature, and its participants. The authors

described how the new ideal of interactivity was having a large impact on how media and

business operated, and would change industries. Downes and McMillan discussed the

nature of the message included in interactivity in terms of its time needed and whether it

was synchronous or asynchronous, how conducive it was to two-way communication

between sender and receiver, and in terms of its place and whether it created a sense of

place. They considered the amount of control the participant had (which is similar to that

described by Heeter in 1989), the responsiveness to the needs of the user, and the goal of

the interactivity. The authors stated that interactivity could be seen as being a continuum

for these components, such that, when sense of place or sense of control increased, so did

amount of interactivity. Researcher stated that even the simplest Website contains some

interactivity as the user has control of what they see through the use of hypertext and

links.









Jensen (1998) also added to the discussion as he described how interactivity should

be looked at in several ways including through psychology, informatics, and mass

communication. He described that interactivity could be a criterion in which something

must be present, or a form of communication offering users the ability to get closer to

interpersonal communication. He also described it as a technology. While this scholar

also discussed the amount of control and effort exerted by the users he agreed with

Downes and McMillan (2000) that interactivity should be looked at as a continuum or as

dimensions. Jensen described four dimensions that range from one-way communication

to n-way communication, which is similar to interpersonal communication and is

continuous. He described four types of communication in interactivity, which includes

registration (two-way with feedback based on what the user inputs), continuous (two way

communication), consultation (pre-defined content a user seeks out with a feedback

loop), and allocution (one-way communication with no feedback). He stated that for

something to be interactive, it must be described by how much effort the user gives

(Jensen, 1998).

McMillan and Hwang (2002) went on to look at interactivity more closely,

focusing specifically on four components. Similar to Jensen, McMillan and Hwang said

that one should look at interactivity through mass communication and consider the

message and the four types of communication (registration, continuous, consultation, and

allocution). As researchers look at this multidimensional concept, McMillan and Hwang

argued that they should consider organizational communication and theorists Gruning

and Gruning's (1989) four-part model. McMillan and Hwang (2002) stated that one must

look at the direction and components of the communication as described by the theorists.









He went on to say that researchers must also look at it in terms of interpersonal

communication and how people look at and use technology, and lastly in terms of media

components and what technology is actually involved.

Liu (2003) attempted to develop a scale to measure the interactivity of Websites.

Three studies conducted resulted in three correlated dimensions of interactivity, including

active control, two-way communication, and synchronicity which comprised a scale to be

applied to marketing and scholarly research.

Researchers have taken the previously discussed descriptions of interactivity to

look further into how interactivity on Websites affects users. Sundar, Kalyanaraman, and

Brown (2003) examined interactivity as a contingency view. They defined interactivity as

a "process involving users, media, and messages, with an emphasis on how messages

relate to one another" (Sundar et al., 2003, p. 35). They conducted an experiment in

which they broke interactivity into three levels: low interactive which contained no links,

medium interactivity which was a single layer of related links, and high interactive which

had two hierarchal layers of links. Results showed that participants viewing the three

conditions did not differ in their ability to recall and recognize content from a site

(Sundar et al., 2003). They did find that the level of interactivity of the site had an

influence on the impression of the political candidate featured on the site. They

concluded that users' perceptions of interactivity were positively associated with the

number of hyperlinks present on the site and the linking actions initiated (Sundar et al.,

2003). Bezjian-Avery, Calder, and lacobucci (1998) also concluded that interactivity may

not always help users. They found in a study of advertising that in certain conditions,

interactivity actually interrupted persuasion as users could move right through the









interactive media without attending to the advertising message. Their study concluded

that the linear traditional advertisements yielded more positive or in some cases similar

results with respect to decisions to purchase. Liu (2003) found that different consumers

may want different levels of interactivity in different situations. Individuals who are "low

interaction-ready" consumers will prefer lower levels of interactivity than those

individuals who are "high interaction-ready" (Liu, 2003).

Sicilia, Ruiz, and Munuera (2005) examined the effect of interactivity on

information processing and favorability toward a product and found contradicting results.

In this study, interactivity was conceptualized as a Website containing six pages

connected through hyperlinks and email, and a non-interactive site set up as a print

advertisement with the message on one page and no hyperlinks (Sicilia et al., 2005). They

concluded that individuals viewing the interactive site processed information more

thoroughly than those viewing the non-interactive site. It was found that motivation to

process the information increased under the interactive condition.

Teo, Oh, Liu, and Wei (2003) investigated the effect of interactivity on attitudes

toward commercial Websites. Results showed that interactivity on a site had a positive

effect on a user's perceived satisfaction and attitude toward a site. They also noted that

interactivity levels significantly influenced the site's effectiveness in helping users in a

decision-making process (Teo et al., 2003). Chung and Zhao (2004) echoed these results

in an experiment testing three levels of site hyperlinking and interactivity. They found

that perceived interactivity had a positive influence on attitude toward the advertisement

and memory of the information on the advertisement. Wu (1999) also looked at

interactivity in advertisements online and concluded that perceived interactivity had a









positive effect on attitude. Wu defined interactivity in terms of responsiveness and

navigability. Chen and Yen (2004) discovered that interactivity on a site is related to

viewers' perceptions of the quality of site design. They suggested that successful sites

should include interactive features that add playfulness, connectedness, and reciprocal

communication.

Linearity

Based on some definitions of interactivity, levels of site hyperlinking can be

considered interactive, thus making it important to look at the literature on linearity.

Several researchers have researched linearity and hyperlinking in Websites (Tremayne,

2004; Lowrey, 2002; Dimitrova, Connolly-Ahern, Williams, Kaid, & Reid, 2003;

Eveland & Dunwoody, 2001; Berger, 2001; Massey, 2004). Dimitrova and colleagues

(2003) found in a study of newspaper Websites focusing on the coverage of Timothy

McVeigh's execution that online papers may use hyperlinks as a gatekeeping device.

Other researchers have looked at linearity and its effects on users. Eveland, Cortese, Park,

and Dunwoody (2004) found through an experiment with college students and adults that

the non-linear structure of the Web has both strengths and weaknesses in terms of

learning when compared to print media or more linear Websites. Researchers also found

that linear site designs encouraged more factual learning while non-linear sites increased

knowledge structure density (Eveland et al, 2004). In contrast, Lowrey (2002) showed

that user recall had no significant difference based on viewing a linear or non-linear site.

Lowrey also explained that linear structure had an effect on the degree of perceived

control over the media, but did not affect the perceived credibility.

Sicilia, Ruiz, and Munuera (2005) studied how consumers processed information

and their experiences with interactive and non-interactive Websites (which they described









as non-linear and linear, respectively). Experimental findings showed that interactive

(non-linear) sites lead to more processing of the information and more favorability

toward the site. Berger (2001) discovered, in an experiment looking at hypertext, that a

significant positive correlation existed between hypertext comfort and user satisfaction.

However, the researcher also reported that hypertext did not correlate with users'

information recall or accuracy in recall. Eveland and Dunwoody (2001) found similar

results in that no significant differences across linear and non-linear Websites were

shown in terms of cued recall.

Adaption/Innovation Theory

One way to look at interactivity and linearity that could explain the differences in

findings could be focusing on how individuals look at the information based on specific

psychometrics like cognitive style. Cognitive style as a theoretical construct is used to

describe and explain an individual's processing of information when solving problems

(Foxall & Bhate, 1993).

Problem Solving

Problem solving has been described as a skill needed for continued existence in

today's technological world (Wu, Custer, & Dyrenfurth, 1996), the means by which life

survives and manages the constant change presented in everyday life (Kirton, 2003).

Problem solving can be seen as the inclination to respond in a certain way when faced

with a problem (Wu et al, 1996). A problem has been defined by Goldsmith and Matherly

(1986) as a situation where standard or customary procedures can not cope with the task

due to unfamiliar elements that encroach. Hedges (1991) described it as a linear process

that begins with users recognizing a specific problem, defining it, having the ability to

comprehend and develop it, test hypotheses and gather data about it, revise those









hypotheses and retest, and then form a conclusion on the problem. The data-gathering

portion of the process is deeply embedded into the usage of communication tools.

As one begins this process, individuals are limited by the way they are built, in

terms of intelligence (Kirton, 2003). Wu and colleagues (1996) found evidence that

differences between technological problem-solving and personal problem-solving style

may exist. Personal problem-solving was defined as problems dealing with depression,

conflict, and life decisions (Wu et al, 1996). The researchers claimed that problem-

solving style is an important difference between individual college students that must be

looked at in terms of students' study of technology.

Kirton's Theory

Researchers have described that all people are bound by their makeup as to how

they define and solve the problems with which they are faced (Kirton, 1999). This

cognitive problem-solving style refers to the characteristic manner of how an individual

will behave over situations and time. This consistent pattern of processing information

influences behavior by organizing it into a system (Foxall & Haskins, 1986).

One way to measure this cognitive style is through the Kirton Adaption-Innovation

Inventory (KAI) (Foxall & Bhate, 1993). This inventory requires respondents to assess

the degree of ease or difficulty they encounter in sustaining adaptive or innovative

behaviors over periods of time. Responses are computed into overall scores ranging from

32 to 160 (Foxall & Haskins, 1986). Respondents who score below the 96 midpoint are

considered adaptorss" while those above 96 are "innovators" (Foxall & Haskins, 1986).

Kirton (2003) states that people can be placed on a continuous scale between being

adaptive and innovative in their problem-solving style. Cognitive style is a trait that can

be expected to be stable over time and across situations (Kirton, 2003) People, however,









may adapt their style through coping behaviors when they find themselves in a particular

situation (Kirton, 1999). For example, one may be seen as being more adaptive at work

and more innovative with friends. Those who are further apart on the KAI scale are more

likely to have problems working together due to these differences; however, when this

happens, people begin to employ coping behaviors so that they are able to avoid these

problems in some situations (Kirton, 2003).

The adoption-innovation theory suggests that adaptors' and innovators' voluntary

styles of cognition differ in three respects: rule conformity, efficiency, and preference for

sufficiency versus proliferation of solutions to problems (Foxall & Hackett, 1992).

Adaptors will solve problems within their existing perceptual frame of reference, while

innovators will change those frameworks and do things differently as they seek solutions

outside of the context of the given problem (Goldsmith, 1984; Foxall & Haskins, 1986).

The KAI can be broken into three sub-scales: sufficiency of originality (SO),

efficiency (E), and rule/group conformity (R) (Kirton, 1999). When looking at the three

subsets of KAI, adaptors and innovators can be described more in-depth. On the

originality scale, adaptors tend to present only a few solutions to problems while the

more innovative person may propose many, possibly impractical, solutions (Bagozzi &

Foxall, 1995). The more adaptive individual will prefer to progress incrementally toward

a goal and an innovator will avoid attention to detail when dealing with efficiency.

Lastly, when comparing the rule governance subset, more innovative individuals ignore

rules or invent their own rules as they go while more adaptive types will prefer to restrict

their behavior to be socially acceptable (Bagozzi & Foxall, 1995). While these subscores









add further insight into how people solve problems, they have been found to be less

reliable with younger populations who are less mature (Kirton, 1999).

Kirton (2003) has described adapters as those who like to have structure in place

when they are attempting to solve a problem. Adaptors may appear cautious as they

prefer to work within an established paradigm of rules (Foxall & Bhate, 1993). They

foresee problem-solving and decision making as a sound, thorough process (Foxall &

Bhate, 1993).

Adaptors are satisfied with devising a small number of sufficient solutions, and

pursue efficiency in problem-solving by making steady progress toward a solution. Their

cognitive behavior tends to be rule-governed in that they prefer to conform as opposed to

break rules (Foxall & Bhate, 1993). Adaptors more readily accept the status quo and will

not challenge the accepted, or try to change, way of traditionally doing things (Kwang,

Ang, Ooi, Shin, Oei, & Leng, 2005). Kirton (1999) also discusses this, saying that they

will work within the established theories, policies and practices. Innovators, in contrast,

are more likely to enjoy a looser structure as they go about solving problems (Kirton,

2003). The innovator will tend to offer more discontinuous solutions to problems while

being seen as adventurous or as a risk taker (Foxall & Bhate, 1993). Innovators have a

tendency to strive for novelty, exploration, trial and error, and risk-taking. Innovators will

promote new understanding through profound procedural changes (Foxall & Bhate,

1993). Innovators are less likely to accept the status quo and do not like to follow the

accepted way of doing things (Kwang et al., 2005). Kwang and colleagues (2005) found

that adaptors and innovators will also prescribe to different values when taking tasks into









consideration. With respect to demographics, KAI scores for females tend to be more

adaptive than males (Foxall & Haskins, 1986).

Foxall and Haskins (1986) suggested that the KAI is a viable marketing tool for

identification of consumer choices. In the area of marketing, researchers have found that

adaptors are attracted to products they know, are less tolerant of change, and are

unwilling to explore (Foxall & Bhate, 1993). They prefer a reasoned argument in

advertising as compared to persuasion (Foxall & Bhate, 1993). Adaptors tend to prefer

the products they currently use and would solve problems that arise from changes in that

product. When searching for information, adaptors will seek and use information

conservatively as they slowly work to a decision (Foxall & Bhate, 1993). Marketing and

business researchers have found that innovators will use more sources of information to

find solutions to a problem, and will trust discrepant advertising, as well as personalized

advertising that encourages them to act impulsively (Foxall & Bhate, 1993). Innovators

will seek information about more innovations than adaptors, even when this information

conflicts with current product use. Adaptors are content with products they currently use

and will not necessarily go looking for new products (Foxall & Bhate, 1993).

Research shows that Adaption-Innovation theory correlates with many personality

traits that are related to consumers, but it also describes relationships between decision-

making and problem-solving in consumers (Foxall & Bhate, 1993). Foxall (1996) found

that the KAI is not a predictor of early adoption of new products. It was also concluded

that innovators would require little personal communication from marketers to adopt a

new product. While adaptors want a lot of reassurance from marketers, it does not matter









if marketers went out of their way to provide this information, as adaptors would

eventually buy the new item either way (Foxall, 1996).

Pershyn (1994), in a study of the KAI on natural creative processes, asked

participants to recall a problem they faced and to draw the process in which they went

about solving the problem successfully. It was found that high adaptors tended to draw a

linear process in which they were orderly working through the process in fewer stages.

High innovators, in contrast, showed a non-linear process in which they had random,

complex approaches with more stages and in some cases no true end point.

The KAI has been utilized in several studies to describe the psychology of

computer usage. Foxall and Bhate (1991) found in a study of graduate business students

that the number of computer applications utilized and duration of computer use were

correlated with KAI scores. The researchers found that for home computer use, those

who were highly innovative tended to use more than four package-based computer

applications and show high personal involvement with computing. Foxall and Bhate

(1991) stated that there is a need for investigations of the relationships with KAI and

computer use. Foxall and Hackett (1992) found in an investigation with managers that

use of software applications was positively related to adaptive-innovative problem-

solving style. They noted that the sufficiency of originality and rule conformity subscales

were positively related to computer use, while efficiency was negatively related to

computer use. This implies that in terms of computer use, sufficiency of originality and

rule conformity are relevant to innovative traits while efficiency is relevant to adaptive

traits (Foxall & Hackett, 1992).









Bhate (1999) used the KAI to examine cognitive styles and different message

sources on attitude change. Through a study of 15- to 73-year-olds, it was concluded that

it was too simplistic for advertisers to use one universal appeal (whether it is positive or

negative) for all individuals. The decision-making process for adaptives was source-

oriented and was influenced by both positive and negative sources. However, innovators

tended to rely more on negative sources as they felt positive sources were more time

consuming. This implies that for information-driven Websites designers need to consider

more than just one type of design for a site.

While the KAI was initially developed for use with adults with work experience, it

has been found to be reliable with younger populations who are also affected similarly by

style (Tefft, 1994; Kirton, 1999). Many university students have had experiences working

that they can pull from when taking the inventory (Kirton, 1999). Dependable KAI scores

have been achieved with people as young as 15 years old; however, due to maturity levels

it has been indicted that the sub scores of SO, E, and R should be ignored (Tefft, 1994).

Taylor (1993) used the KAI for a group of 17- to 18-year olds and noted that the theory

worked with this younger audience in a similar way as it does for adult populations.

Taylor did discuss the need for explanation of a few words in the KAI that were

confusing to this youthful audience. Fisher, Macrosson, and Wong (1998) reported

successfully using the KAI with undergraduate students in engineering and business to

test relationships between cognitive style and team role preference. Foxall (1992) also

reported successfully utilizing the KAI with a group of students enrolled in masters of

business administration programs in the United Kingdom, Australia, and the United

States.









Information Richness

The concept of information richness comes into play as people cognitively choose

which media will help them as they work to solve problems. Individuals choose media

they perceive to be most efficient in helping them to complete a communication task

(Kelleher, 2001). Based on equivocality or the ambiguity, the lack of clarity of

information, users who are seeking information will choose either a "rich" or "lean"

medium (Irani & Kelleher, 1997). Equivocality can be described as the existence of

multiple interpretations and an organizational situation (Trevino, Daft, & Lengel, 1990).

Trevino and colleagues further described equivocality by comparing an equivocal

message to a Rorschach ink-blot-multiple users may read it differently than others

depending on their unique backgrounds and perspectives. Information tasks that are seen

as unambiguous will be considered low in equivocality, while tasks that are based on

processing of multiple interpretations may cause higher equivocality. The theory of

information richness describes that users use rational media choices to deal with these

equivocalities (Irani & Kelleher, 1997). Equviocality and ambiguity of a site can be tied

back to many of the same underpinnings that drive adaptors and innovators. This idea of

something being equivocal or ambiguous or rich or lean is similar to the way the KAI

discusses structure. While more adaptive problem-solvers will like structure, it could be

argued that they too may like equivocal messages that are "lean," and more innovative

problem-solvers who work outside structure will want more ambiguity in their messages

that are "rich."

Media richness refers to a medium's tendency to present information in either a

"rich" or "lean" manner. The richest communication medium is face-to-face, followed by

the telephone and e-mail because these media allow immediate feedback and can be









highly personal (Trevino, Daft, & Lengel, 1990). The leanest communication is formal

documents such as fliers, bulletins, and quantitative reports. Irani and Kelleher (1997)

found in a study of equivocality that those in a high equivocality condition will choose a

richer Website more often than a pamphlet or a lean Website. Based on this perspective,

it could be assumed that as people work through the cognitive processing of their

information-seeking tasks, not only their preferred style will come into play, but also the

complexity of the task will influence the media choice.

Uses and Gratifications Theory

Uses and gratifications theory also addresses cognitive style as a motivator to fulfill

a specific need (Stone, Singletary, & Richmond 1999, p. 200) based on an individual's

choice of media. The theory emerged in the late 1950s and early 1960s as researchers

looked to understand audience involvement in mass communication (Blumler, 1979). The

Uses and Gratifications theory provided a replacement of the ideal that the audience

member was a passive victim, and that one could actively look at media for their own

purpose (Blumler, 1979).

As one of the theories most associated with usage of media, Uses and Gratifications

researchers define communication needs that shape why people use media and the

behaviors that gratify those needs (Rubin, 1994). One such use that is noted in the

literature is to find information or solve a problem (Katz, Blumler, & Gurevitch, 1974).

Uses and gratifications theorists assume that communication needs interact with social

and psychological factors to produce motivation to communicate (Rosengren, 1974).

Katz, Blumler, and Gurevitch explain that the logical steps the theory is concerned with

include "(1) the social and psychological origins of (2) needs, which generate (3)

expectations of (4) the mass media or other sources, which lead to (5) differential patterns









of media exposure (or engagement in other activities, resulting in (6) need gratifications

and (7) other consequences, perhaps mostly unintended ones" (1974, p. 20). Five

elements that have been described as assumptions to Uses and Gratifications research

include 1) active audience, 2) audience member links need of gratification and media

choice, 3) media compete with other sources for need satisfaction, 4) goals of mass media

use can be derived from data provided by individual audience members, and 5) value

judgments about mass communication need to be suspended when audience orientations

are explored (Katz, Blumler, & Gurevitch, 1974). Other assumptions associated with

Uses and Gratifications approaches include a) media use is goal directed, b) media

consumption can fill a wide range of needs, c) people have enough self-awareness to

know and articulate their reasons for using the media, and d) gratifications have their

origins in media content (McLeod & Becker, 1974).

From a Uses and Gratifications standpoint the first assumption is that media users

are active in their attempts to seek out and find information from the media channel of

their choice (Rubin, 1994). The approach assumes that users are active participants

because they are active communicators who select which communication channel to use

(Blumler, 1979). Theorists suggest that users will have some form of need, such as

solving a problem, which they will try to gratify with the use of a specific media (Baran

& Davis, 2002). These motives and needs can be based on psychological characteristics,

attitudes, and perceptions (Rubin, 1994).

As seen in the second assumption put forth by Katz, Blumler, and Gurevitch

(1974), media fulfills one of four functions for individuals: it entertains them, serves as a

mechanism for surveillance, correlates with what people know about society; or transmits









society across generations (Baran & Davis, 2003). Based on these functions, media users

will feel a need (needing to solve a problem, needing love or acceptance, or needing to be

informed or entertained) that they will be motivated to gratify through a specific media

outlet (Blumler, 1979). These needs lead someone to actively seek out and use a specific

medium that will in turn gratify that perceived need (Bran & Davis, 2003). The medium

individuals chose would be based on their expectations of that medium and their

perceptions of how well it will gratify that need (Blumer, 1979). This theory assumes that

communication behavior is sought out to fulfill these cognitive needs by an individual

user (Katz et al., 1974).

The third assumption calls for researchers to realize that media compete for the

ability to fulfill user needs (Katz, Blumler, & Gurevitch, 1974). Those needs served

through the media and communication is only a small segment of the human needs that

need fulfillment, and the degree to which those can be fulfilled through media varies

(Katz, Blumler, & Gurevitch, 1974). Bouwman and Wijngaert (2002) called for

researchers to take into account the personal factors and situations that affect media

choice, as was traditionally called for in this Uses and Gratifications approach.

Uses and Gratifications and the Internet

The Uses and Gratifications paradigm has been described as the best model in

which to study new communication methods such as the Internet (Ruggiero, 2000). Many

recent studies have utilized Uses and Gratifications as part of the theoretical framework

when studying the Internet because of the interactivity, demassification, and

asynchronicity it allows that other media do not (Baran & Davis, 2003). Such studies

have shown that Internet use is motivated by the need to escape, the need for

entertainment, the need for interaction, and the need for learning and socialization (Baran









& Davis, 2003). Papacharissi and Rubin (2000) examined audience uses of the Internet

and identified five motives for using the Internet including: information seeking,

interpersonal utility, pass time, convenience, entertainment. Their findings also suggested

that those who like to look around the Internet felt it allowed them to save money and

obtain information (Papacharissi & Rubin, 2000). Through several case studies Bouwman

and Wijngaert (2002) concluded that due to the fact that the receiver of information on

the Internet has to seek it out, the assumption of an active audience in the Uses and

Gratifications approach is greatly supported.

Ko (2002) investigated whether motivations to use the Internet could explain key

aspects of usage. Information, escape, time passage, and interactivity were the four

primary motivations to using the Internet discovered by the researcher. Ko (2001) found

that those who use the Internet for information are more likely to satisfy their needs by

using the Internet over other media. Web users are active information seekers, as they

must click on links and use hypertext to navigate online. Lin and Jeffers (1998) suggested

that, in turn, Web use is goal directed as the users must be aware of the needs they

attempt to satisfy. Ko, Cho, and Roberts (2005) looked at the Internet use of college

students in the United States and Korea through an experimental design. Researchers

concluded that consumers who possessed high information motives were more likely to

engage in human-message interaction on a Website; while those with higher social

interaction motives were more likely to engage in human-human interaction.

The Uses and Gratifications approach has also been utilized to study political

information posted online. Kaye and Johnson (2002) found that the four primary

motivations for locating political information online included guidance, information









seeking/surveillance, entertainment, and social utility. It was found that guidance and

information seeking/surveillance is linked to more purposeful uses of the Web than for

just surfing.

Previous Experience and Expectations

As individuals choose the media that they will utilize to gratify their needs, they

draw on memories of past media use to aid them in that action (Katz, Blumler, &

Gurevitch, 1974). The third step in the model, expectations, has been has been thoroughly

researched through the Uses and Gratifications paradigm (Rayburn, 1996). Katz and

colleagues (1974) allude to the expectations from media by the audience when selecting

content to fulfill certain needs. The medium that is used depends on a variety of factors,

including the characteristics of the information needed, the characteristics of the person

asking the question, and the context in which they have access to specific media

(Bouwman & Van De Wijngaert, 2002).

Peled and Katz (1974) found in a study of media during wartime and crisis that

people came to a specific medium with expectations of what that medium will be and

what it will gratify for them. Bouwman and Wijngaert (2002) found in a study of

characteristics of basic needs that certain thresholds of accessibility must be met before

deciding to use a medium. One such threshold they describe is that of suitability, where

the medium can provide them the information they are searching for (Bouwman &

Wijngaert, 2002). For the user to know if that medium has that information, they must

have some previous experience with it.

Several theories have attempted to deal with theses expectations, such as Fishbein

and Ajzen's (1975) expectancy-value theory. This theory states that there are three kinds

of beliefs: descriptive which is a result of direct observation; information which is formed









by accepting information from outside sources; and internal which include characteristics

of the object that are not directly observed (Rayburn, 1996). This model has shown to be

of great use to the study of Uses and Gratifications where feedback loops are prevalent

(Rayburn, 1996).

Information Recall

The Internet may not always be the best medium to reach audiences; Bogart (2000)

reported that an experiment done at The Ohio State University showed that when readers

were given an article in both print and Web versions, they reported that the printed

version was more understandable. "A strength of the Web is its ability to present

individual readers with a selection of tailored contents. This is also a weakness, if it

means that they are no longer exposed to what they have not expected and did not know

they wanted," (Bogart, 2000, p. 1).

In contrast, D'Haenens, Jankowski, and Heuvelman (2004) found in their study of

two online and print versions of Netherlands newspapers that there was no evidence that

online readers consume and retain news differently than those reading print versions.

They found that online readers recalled international news better than print readers. It was

concluded that no evidence in the study supported claims that online readers consume

news differently from print readers. Moore (2004) found in an experimental study of

magazine and online advertisements that higher selective exposure was found for

information online over the print version, and moderate recall differences were seen

between the two media. Based on these findings, Moore called for future research of new

media to examine memory and media comparisons.

Eveland and Dunwoody (2002) found that when compared to print media, the Web

increases learning through an increased elaboration, but may decrease it through









increased selective scanning. Tewksbury and Althaus (2000) found similar results with

people reading online newspapers versus print additions. It was found that the online

versions of the papers presented fewer clues to the importance of events compared to

print editions, and in turn people were more willing to use their own interest to guide

what they focused on and were able to recall. Danaher and Mullarkey (2003) found that

length of exposure to a site containing a banner advertisement affected how likely

viewers were to be able to recall the information. It was suggested by the researchers that

designers should include interactive features that encourage users to stay on a page

longer. It was also noted that those in a "goal-directed mode" were less likely to

remember information than those who were just surfing the Internet. While Till and

Baack (2005) did not look at recall in terms of information online, they did add to the

discussion by looking at the creativity of advertisements in terms of recall. It was

discovered that in an unaided basis, creativity generates significantly more recall.

Eveland and Dunwoody (2001) compared learning in print versus linear, nonlinear,

and advisement Web designs. It was found that learning was better for print than

nonlinear and linear; however, no difference was found between print and an

advertisement design (which included cues to work through the site such as "back,"

"next," and "story map" buttons). It was also noted that Web experts learned more than

Web novices on all mediums.

Wicks (1995) used an experiment looking at free recall, or recall not prompted, and

extended recall, or recall after time has passed, to see the effects of medium on news

recall. Wicks found that individuals acquire "common knowledge" from the news and

that time is needed in the recall process.









Several studies compared different forms of online media and other media to

discover more about recall of information. Berger (2001) ascertained that those

comfortable with hypertext did not have a significant difference in recall than those with

low comfort levels, concluding that presenting information linearly or nonlinearly would

not offer users an advantage. Lowery (2002), however, found that linearity has an effect

on perceived control over the media experience, but did not lead to any increased

knowledge. Fox and colleagues (2004) found in a study of television news that recall was

greater for younger and older viewers when graphics were present. Multimedia, such as

video and imagery, has also been found to not increase comprehension or recall scores

above those of a static, text-based site (Berry, 2001). Berry (2001) felt that multimedia

may enhance a user's recall of textual information if it was reinforcing the textual

information on the page.

Eveland and colleagues (2004) utilized an experimental design to discover how

Web site organization influenced free recall of information. The researchers concluded

that a nonlinear compared to a linear site had mixed results on learning. It was found that

a linear site design increased factual learning in participants, but the nonlinear design

increased the knowledge structure density. The learning of factual knowledge was

hindered by the nonlinear structure.

Attitude

Attitude has been shown to be an important predictor of usage and implementation

of technology (Rodgers & Chen, 2002). Eagly and Chaiken (1993) described attitudes as

being derived to motivate behavior in order to exert effects at various stages of

information processing. It is further conceptually defined by the authors as "a

psychological tendency that is expressed by evaluating a particular entity with some









degree of favor or disfavor" (Eagly & Chaiken, 1993, p 1). An attitude is not formed until

people are presented with a situation in which they must evaluate it on an effective,

cognitive, or behavioral basis (Eagly & Chaiken, 1993). While attitudes are not directly

observable, they can be inferred from responses given that show some state or disposition

that has been engaged (Eagly & Chaiken, 1993).

Researchers have assumed that attitudes should be divided into three classes -

cognitive, affective, and behavioral (Eagly & Chaiken, 1993). The cognitive category

contains all of the thoughts an individual has about the attitude object, while the affective

category is the feelings and emotions one has in relation to the attitude object (Eagly &

Chaiken, 1993). The behavioral category contains one's actions with respect to the

attitude object. In congruence with the idea of three categories of attitudinal responses, is

the idea that there are three antecedents to attitude cognitive processes, affective

processes, and behavioral processes (Eagly & Chaiken, 1993).

The cognitive process in which attitudes draw from is one in which much research

derives (Eagly & Chaiken, 1993). The assumption by researchers is that attitude is

derived from a cognitive learning process in which one gains information about the

attitude object and then forms beliefs. The information is gained via direct and indirect

experiences with the attitude object (Eagly & Chaiken, 1993).

While research on attitudes has been more defined in the social psychology

literature, it has been found in other social science literature as well (Eagly & Chaiken,

1993). Attitude research has been done extensively in the area of advertising and attitude

toward ads and their effectiveness (Chen, 1999; Rodgers & Chen, 2002; Sinclair & Irani,

2005; MacKenzie & Lutz, 1989). Rodgers and Chen (2002) found adoption of the









Internet by advertising agencies is affected by poor attitudes toward and lack of

experience with Internet advertising. In 1999, Chen developed a scale, based on other

evaluative scales, to provide researchers the ability to measure the attitudes of users about

Websites to help indicate the value of such sites.

Cho (1999) found in a study of advertising on the Web that people who had more

favorable attitudes toward the Web were more likely to click on advertising on a site.

Rodgers and Chen (2002) looked at advertising in terms of the organization, and found

that poor attitudes toward the Internet after adoption for advertising was due to the

agencies' lack of experience and expertise with that form of advertising. Teo and

colleagues (2003) found that attitude toward a commercial Website can be positively

influenced with increased interactivity on the site.

Conceptual Framework

Based on the literature presented in this chapter, the conceptual framework seeks to

explain a model in which: An individuals' cognitive problem-solving style, when

influenced by their level of previous usage of the media will affect their levels of attitude

and information recall after being presented with an interactive or non-interactive

Website (Figure 2-1.).










Recall Attitude
Level of High High
Usage



Problem Solving Level of
Style Interactivity of
(More Adaptor/Innovator) Site



Recall Attitude
Low Low




Figure 2-1. Conceptual framework for this study.


Research Questions

Based on the conceptual framework, this study will examine the relationship of

cognitive problem-solving styles and level of Internet usage on perceived attitudes

toward and information recall of sites that vary in their level of interactivity.

This study will attempt to answer the following questions:

1. To what extent does problem-solving style and Internet usage have in influencing
perceptions of attitude and recall toward Websites that vary in level of
interactivity?

2. To what extent is problem-solving style alone a factor in influencing perceptions of
attitude toward and information recall toward such a site?

3. To what extent is Internet usage alone a factor in influencing perceptions of attitude
and level of information recall toward such a site?














CHAPTER 3
METHODOLOGY

Overview

As Extension professionals continue to embrace online technologies to reach

audiences and inform them about topics that will help solve problems with which they are

faced, it is important that this information be presented in a form that will be usable,

valuable, appropriate, and easy to recall. This method of disseminating Extension

information is providing new challenges, as it is attempting to reach audiences in a new

format that is usable and valuable.

Thus, the purpose of this investigation was to examine the effects of individuals'

problem-solving style, level of Website interactivity, and level of Internet usage on

attitude and level of recall with respect to an information-driven Extension Website.

Although usage has been looked at extensively in the literature, it has never been tied to

problem-solving style. By understanding how problem-solving styles in particular affect

users' perceptions of a Website with respect to such attributes as attitude and information

recall, Extension, agricultural communicators, and commodity groups who are utilizing

the Internet to reach audiences will be better able to utilize communications processes to

inform audiences, educate them, and affect productive change.

Hypotheses

Based on the literature presented, the following hypotheses were developed:

H1: Unaided information recall and level of attitude toward an information-driven
Extension Website will differ significantly as a function of site interactivity,
problem-solving style, and level of Internet usage.









For Subjects Who Receive the Interactive Site:

H2: Attitude toward an information-driven Extension Website will differ
significantly as a function of Internet usage and problem-solving style.

H3: Unaided information recall will differ as a function of Internet usage and
problem-solving style.

For Subjects Who Receive the Non-Interactive Site:

H4: Attitude toward an information-driven Extension Website will differ
significantly as a function of Internet usage and problem-solving style.

H5: Unaided information recall will differ as a function of Internet usage and
problem-solving style.

Research Design

The research design for this study was experimental in nature. The design was a 2

(more adaptive/more innovative problem-solving styles) x 2 (interactive/non-interactive

information-driven Extension Website) x 2 (high/low levels of Internet usage) between-

subjects, factorial design focused on assessing whether the problem-solving style of an

individual, coupled with exposure to an interactive or non-interactive information-driven

Extension Website and level of Internet usage, will influence information recall and

attitude.

Factorial designs allow for a more significant test of hypothesis (Ary, Jacobs, &

Razavieh, 2002) by determining the influence of an individual independent variable on

another independent variable (Christenson, 2001). Factorial designs not only offer the

ability to test more than one independent variable, but they also allow for the testing of

more than one hypothesis in one experiment (Christenson, 2001). Beyond testing just the

independent influence of a specific variable, factorial design allows for tests of

interaction to be performed (Christenson, 2001). In order to generate a large enough

sample to effectively test the hypotheses, a minimum sample size of 180 (2 x 2 x 2 x 30=









240) is needed to ensure at least 30 subjects for each condition (Christensen, 2001). The

group design appears as follows (Gall, Borg, & Gall, 1996):

R X1 O1

R X2 01

In this posttest-only, randomized subject design: R= random assignment, O1=

posttest measures, X1= interactive version of the site, X2= non-interactive version of the

site.

The basic threats to internal validity were considered in the design of the study. As

identified by Campbell and Stanley (1963), threats to be cognizant of include history,

maturation, testing, instrumentation, regression, subject selection, mortality, and

interaction effects. The post-test only design of this study was expected to address

regression, history, and maturation (Campbell & Stanley, 1963). Mortality was a concern

in this study as students were asked to give data on two different occasions, once in the

classroom and once online. Time between these two administrations was less than a day.

To address this concern, data was examined post collection, and showed that this was not

an issue since only 10 participants did not complete both sections. Interaction of

extraneous variables was also a possibility. To address this concern factorial design was

utilized. A factorial design allows for the confounding variables to be built into the

design. Confounding variables that were controlled for included gender (Kirton, 2003),

time, content (Saunder, 2000), and previous experience. To address instrumentation

validity, a panel of experts was utilized to look at the face validity of the items, and a

pilot test was run to ensure construct validity. The pilot study included testing of the

instrument as well as the message stimulus. Threats due to testing are described as the









effects of taking a first test upon the scores of a second test (Campbell & Stanley, 1963).

This could be a concern as the participants took two different instruments at different

times. However, the instruments were on different topics and were not in a pre-test/post-

test situation where they would affect one another.

While the undergraduate courses, in which the sample was derived, were selected

through availability, to counteract any validity threats posed by the selection of

participants, the version of the sites were randomly assigned to the participants and data

was collected on each individual student and not the course as a whole. Manipulation

checks were conducted on the version of the sites to ensure no differences were found.

These manipulation checks will be described in more detail later in this chapter.

Threats to external validity were also taken into consideration during the design of the

experiment. Threats which need to be addressed, as outlined by Ary and colleagues

(2002), include population validity, ecological validity, and experimenter validity

operation.

The threat of population validity must be taken into account as the population

consisted of students in three undergraduate courses at a large Southeastern land-grant

university. While this sample is not generalizable to the whole population, the majority of

Internet users are in the 18- to 30-year age range (CyberStats, 2005) and are a viable

population to study in terms of Internet modality. Students' familiarity with the Internet

aids in assuring differences between exposure groups is unlikely due to the novelty

inexperience of the Internet (Tewksbury & Althaus, 2000). Bull and colleagues in 2004,

called for Extension to pay specific attention to underserved audiences who have

typically not been an original stakeholder in the program.









Ecological validity describes how generalizable the experimental environment is to

other environments by taking into account pretest and post-test sensations, multi

treatments, Hawthorne effects, novelty effects, and experimenter effects (Ary et al.,

2002). To address these threats, students were given science-generated content that is

viable and interesting to them. Questions were asked to ensure this topic was of interest

during the study. To help ensure a more natural environment when taking the instrument,

students were asked to take the second portion of the study at home on their own

computers. By doing this, subjects were in a familiar setting and will be looking at a topic

in which they might research on their own time. This also helped to curb any novelty

effects that could occur during the treatment.

Subjects

Participants were recruited out of two service courses taught in the College of

Agricultural and Life Sciences at a large Southeastern land-grant University. The courses

serve as part of a general education requirement for students across the university and are

thus taken to be largely representative student population with a variety of majors and

backgrounds. A total of 314 students completed the initial usage and problem-solving

instruments through direct administration. Those students who were enrolled in more

than one of the courses utilized in the study were instructed to participate in the study

only once. Their names were noted during the first data collection to insure that they did

only participate once. In order to manipulate the treatment version of the site, subjects

were randomly assigned to either an interactive or non-interactive version of the same

Extension Website. Manipulation checks were completed during the pilot study to insure

that there were no significant differences among the two courses or those receiving one

version of the site over the other (Table 3-1).









Table 3-1. Independent Sample T-test for Significant Differences Between Courses and
Version of the Site Based on Age or Gender.
T df Sig. Mean Diff.
Courses
Gender .80 253 .43 .26
Age -.79 254 .43 -.06
Version of the site
Gender 1.27 253 .20 .41
Age 1.1 254 .28 .08


Pilot Study

To establish reliability and validity of the final instrument, a pilot study was

conducted with 29 undergraduate students in an agricultural leadership course at a large

Southeastern university. Care was taken to ensure no students who participated in the

pilot were a part of the final sample. The same procedures were utilized in the pilot study

as they were for the full study, as outlined below. Prior to the pilot test, a panel of experts

assessed the face and content validity of the instrument. To assess construct validity, item

analysis was run on the pilot instrument. An overall Cronbach's alpha reliability of .72

was computed. A few adjustments were made to refine and finalize the instrument.

Message testing was also completed during the pilot study. Participants in the pilot

section were cross-referenced with participants in the full study via the reported emails to

ensure no subject participated in both data collections.

Manipulation checks were conducted during the pilot study to assess whether

respondents could distinguish between the treatment and control version of the Websites.

Subjects in the treatment and control version of the sites were asked to identify if the

version of the Website they were presented with was interactive. An independent-sample

t-test was conducted using version of the site as the independent variable. Results show

that participants receiving the interactive version of the Website were successfully able to









identify that the version of the site they viewed was interactive. Those receiving the non-

interactive version of the site were also able to identify that their version of the site was

non-interactive. A significant difference was found in the response between the condition

groups. Their response is presented in Table 3-2 and Table 3-3.

Table 3-2. Means Table for Website Interactivity Identification (N=19*).
Page is Interactive N Mean SD

Interactive Version 10 3.00 1.50
Non-interactive Version 9 1.44 .73
*19 out of 29 pilot study participants completed the manipulation check. Using a 1-5
Likert Scale (1= Strongly Disagree and 5= Strongly Agree)


Table 3-3. Independent Sample T-test for Significant Differences Between Less and
Interactive Version of the sites.
T df Sig. Mean Diff.
Non-interactive/interactive Version -2.84 17 .011 -1.56


Procedure

The instruments for both the pilot and full studies were administered in two parts

due to the length of instruments and the use of a standardized instrument (KAI), which

could only be administered on paper. The first part of the instrumentation was

administered in the classroom. During this time a certified KAI representative

administered the KAI to the students.1 Participants were also given questions on basic

Web and media use, attitudinal scales on perceptions of the Internet, and demographics.

Upon completing the first part of the instrumentation, participants were asked to report a

university email. Students were informed that they would be contacted later that day via

email with the second part of the instrumentation. They were informed that the course

1 Dr. M.J. Kirton, director and founder of the Occupational Research Center, developed this psychometric
inventory. KAI administrators must complete an intensive weeklong course given by Dr. Kirton on the KAI
instrument and the underlying theory in order to be certified to administer the instrument.









title would serve as the subject to the email, to ensure students did not pass the email off

as spam.

Once students were randomly assigned to either the treatment or control, part two

containing a link to the appropriate version of the site was emailed (Appendix B).

Participants could only view one of the two versions based on the random assignment.

Once at the appropriate version of the site, participants viewed a consent screen and were

directed to click on a link to open a new browser window containing the information

page of the site (Figure 3-1). Participants were instructed to spend as much time as

needed to review the content before completing the final instrument.


t -' -* : -Su."p 2.a -- .














Figure 3-1. A screen capture of the consent information and instructions sent to
participants.

After exposure to the version of the site, participants were instructed to close the

browser window containing the information page and not to return to it. They were then


directed to click on a link which took them to a new page containing part two of the

instrument in which they were asked their attitudes toward the treatment or control

version of the site to which they were exposed, were asked to recall information and

report knowledge on the topic presented, and were asked about their current and past









usage of Extension information through an online form. After data collection was

completed, the data from the part one instrument and part two instrument were matched

based on an email identifier, which was included in both sets of instruments.

The final subjects were solicited out of two large service courses taught at a large

Southeastern university (N=314). A total of 314 students completed part one of the

instrumentation in the class. A total of N=305 individuals completed both parts of the

study. One of the instruments was then thrown out for being incomplete. Another 48

instruments were thrown out due to the sensitivity of the KAI instrument, leaving 256

total participants for the full study.2

Instrumentation

Instruments for the first part of the study consisted of the Kirton Adaption-

Innovation Inventory (32 items), a 49-item instrument measuring media usage, (7 items)

Internet experience (17 items), attitude scales toward the Internet (11 items), a scale on

the value of the Internet (8 items), and demographics (6 items). (See Appendix A.) The

instrument in the second part consisted of 34 items measuring attitudes toward the

interactive or non-interactive version of the site to which they were exposed (11 items),

information recall (4 items), Extension usage (2 items), and knowledge and interest in the

car-buying information (10 items). (See Appendix B.)






2 The KAI is a very sensitive instrument in which participants who respond that everything is easy or hard
for them or those who select down the middle of the scale must be rejected as it is suspected that they are
being reluctant to respond truthfully or are trying to deliberately score in an "acceptable" way (Kirton,
1999, pp.19). It is noted by KAI researchers (Tefft, 1994; Kirton, 1999) that younger populations will have
lower maturity levels, affecting the rejection rate of such a population. However many university students
have enough work experience to understand the items without problems (Kirton, 1999). A 10%-20%
wastage rate can be expected under favorable conditions with university students (Kirton, 1999).







57


Independent Variables

Treatment

The only independent variable that was manipulated in this study was the level of

Webpage interactivity to which the subjects were exposed. The other independent

variables were measured on the basis of data collected via instrumentation.

For the purposes of this study, two versions of a Webpage were created: an

interactive version and a non-interactive version (Appendix C.). The versions were

created using the same information; only the ability to interact with the message was

manipulated (Sicilia, Ruiz, & Munuera, 2005). The non-interactive version was set up

like a traditional Extension fact sheet, that are typically made available in PDF or basic

HTML formats online, where the entire message was on one Webpage with no interactive

elements (Figure 3-2). The Extension logo was visible, and bolded headlines broke up the

text.


ite rs 1 ii








Ordnine a Car

JPu 4r fim l trr sii i l hart traMaimT lra"teyd ihal |yir d cu ol my

1t IE l- I 1X1. I1 -J i 1 M I

L-i- M d .Pm *I:tl .c .IT




Figure 3-2. A screen capture of the non-interactive control page sent to participants.
tI*s TI* 4lit*1A ~Kh Rt i D uici7 biJll^ <^^i.^W^1^^rie^,lftL^











Figure 3-2. A screen capture of the non-interactive control page sent to participants.










The interactive version contained the same information and Extension logo

presented in a Macromedia Flash format (Figure 3-3). After the Extension logo an

animated car could be seen moving across the screen. An introduction and instructions to

click along the "car-buying path" followed. Users could then click on the icons along the

"car-buying path" and a pop-up window would appear with information that could be

moved around the screen or closed.



I .I h I ...... .". 1 a t".... Q *'V *_* |rr

IUNg[YE|RSMT OFm
FLORIDA S[ON
IPAS EXJ ENSION Edwin ON


jiCJ l1


A -, 4imDJ


Figure 3-3. A screen capture of the interactive version of the site page sent to
participants.

The information content of the page was consistent across both stimuli conditions.

The information used was adapted from an Extension fact sheet produced by the

University of Iowa (1998). To make it relevant and salient to the subjects, the topic of car

purchasing was chosen as the focus of the information presented on the page used for the









treatment and control version of the sites. This was due to expectation that the

information would be salient to this audience. College students are in a unique situation

in that they are becoming more important to automakers as a demographic which will

soon be making car purchases (Clemens, 2005; Collier, 2006). These "echo boomers" are

at the prime age to buy their first car and are said to spend more of their income on

products, such as automobiles, than others (Clemens, 2005). Market research has shown

that these young, first-time buyers are turning to the Internet as a main source when

making car-buying decisions (Associated Press, 2006). As individuals begin looking at

buying large items like a car they will move into a problem-solving mentality of trying to

decide what type of car is best for their needs.

Problem-solving style

As described in Chapter 2, all people are bound by their makeup as to how they

define and solve the problems with which they are faced (Kirton, 1999). In this study, the

Kirton Adaption-Innovation Inventory (KAI) was utilized to place individuals on a

continuous scale between being adaptive and innovative in their problem-solving style.

This cognitive style is a trait that can be expected to be stable over time and across

situations (Kirton, 2003). It was assumed problem-solving style would influence

preference for and recall of information needed to solve a problem, such as how to

purchase a car.

The KAI inventory requires respondents to assess the degree of ease or difficulty

they encounter in sustaining their adaptive and innovative behaviors over periods of time

by drawing an "x" where they fit in a series of 32 five-point scaled items ranging from

"very easy" to "very hard" (Foxall & Bhate, 1993) (See Table 3-4). Individual scores are

composed of three independent sub-scales which measure originality (13 items),









efficiency (seven items), and rule-conformity (12 items) (Goldsmith, 1984). Responses

are computed into overall scores ranging from 32 to 160 (Foxall & Haskins, 1986).

Respondents scoring below the 96 mid-point are considered adaptorss," while those

above 96 are "innovators" (Foxall & Haskins, 1986). The KAI inventory has been shown

to have a high level of internal consistency; Kirton (1999) returned a Kuder-Richardson

20 reliability of .88 and then retested with a similar population a year later to again

receive a K-R 20 of .88. In The KAI Manual it is reported that 31 studies from 12

countries have yielded Cronbach's alpha's ranging from .79 to .91 (Kirton, 1999).

Goldsmith (1984) reported a reliability Cronbach's alpha of .84 for the KAI while

Goldsmith and Matherly (1986) reported a Cronbach's alpha of .87. Both studies were

run with undergraduate populations.

Table 3-4. Example of KAI Instrument Items
How easy or difficult do you find it to present yourself, consistently, over a long
period of time as:
Very Hard Hard Easy Very Easy
A person who is patient
A person who conforms
A person who enjoys the
detailed work.


Internet Usage

As discussed in Chapter 2, communication needs interact with social and

psychological factors to produce motivation to communicate (Rosengren, 1974). These

factors and previous experiences with a media will influence a user to choose a specific

media to gain information. Ko (2001) found that those who use the Internet for

information are more likely to satisfy their needs by using the Internet over other media.

Web users are active information seekers, as they must click on links and use hypertext to

navigate online. College students represent the largest population of Internet users







61


(Eastin, 2001). Large percentages, 96%, of all 18- to 29-year-old users find the Internet a

good way to get information (Fallows, 2004). Level of Internet usage is thus important to

understand and gauge. For the purposes of this study, Internet usage was defined by the

amount of Internet use each week, the number of sites subjects visit, and the activities

they perform while online. In this study level of usage was measured through a 13-item

researcher-developed scale (Appendix A).

In order to measure subjects' usage of and experience with the Internet, subjects

were asked several researcher-developed questions about how many hours they spend

online each day and how many sites they visit in an average session. Respondents were

asked to rank on a five-point Likert scale how often they participate in 10 specific online

activities such as downloading music and shopping online (Table 3-5).

Table 3-5. Example of Internet Usage Items.
Please indicate how often you do the following each week
Download music 1 2 3 4 5
Never Sometimes Very Often
Read a blog 1 2 3 4 5
Never Sometimes Very Often
Instant message 1 2 3 4 5
Never Sometimes Very Often
Read Facebook or MySpace 1 2 3 4 5
Never Sometimes Very Often
Watch videos 1 2 3 4 5
Never Sometimes Very Often
Download RSS 1 2 3 4 5
(Real Simple Syndication) Never Sometimes Very Often
Shop online 1 2 3 4 5
Never Sometimes Very Often
Shop/sell on EBay 1 2 3 4 5
Never Sometimes Very Often
Use a search engine 1 2 3 4 5
Never Sometimes Very Often
Work on WebCT or other online 1 2 3 4 5
course Never Sometimes Very Often
How often do you use the Internet to 1 2 3 4 5
find news/information online Never Sometimes Very Often
How many hours a week do you 1 or 2-3 4-5 6-7 8-or more
spend on the Internet less
How many sites do you visit on an 1-2 3-4 5-6 7-8 9 or more
average session online











Dependent Variables

Information Recall

Several researchers have looked at how different components and interfaces online

affect information recall; however, after a thorough literature review, no studies were

found that have examined how this is affected by trait variables such as cognitive

problem-solving. Danaher and Mullarkey (2003) have found that length of exposure to a

Website containing a banner advertisement affected how likely viewers were able to

recall the information. Berger (2001) discovered that those subjects comfortable with

hypertext did not differ significantly in recall than those with low comfort levels. Lowery

(2002) found that linearity has an effect on perceived control over the media experience,

but did not lead to any increased knowledge.

Information recall has been typically measured by asking participants to engage in

free or unaided recall followed by a set of aided recall questions (D'Haenens, Jankowski,

& Heuvelman, 2004; Davis et al., 2005; Danaher & Mullarkey, 2003). Strong

correlations have been reported by researchers utilizing both unaided and aided recall

(Davis et al., 2005). Other researchers comparing the differences between free or unaided

and cued or aided recall found no significant differences in the findings (Padilla-Walker

& Poole, 2002). However, while they are related, it has been noted in psychology

literature that they represent different tasks (Padilla-Walker & Poole, 2002). It has been

discovered that aided recall can cause more recollection of weaker memories, thus giving

less accurate results (Padilla-Walker & Poole, 2002). For the purpose of this study, the

learning process and free recall of information is of interest, rather than recognition.

Thus, participants were asked, after reviewing the interactive or non-interactive version









of the message structure to which they were exposed, to recall information through an

unaided response where they listed all information recalled from the site (Davis et al.,

2005; Eveland, Cortese, Park, & Dunwoody, 2004; D'Haenens, Jankowski, &

Heuvelman, 2004). All true statements were scored as a +1 while untrue statements were

coded as a -1 (Davis, et al, 2005). Points were summed to attain a mean unaided recall for

the information contained in the version of the site to which they were exposed.

Attitude

Attitude has been shown to be an important predictor of usage and implementation

of technology (Rodgers & Chen, 2002). While attitudes are not directly observable they

can be inferred from responses given that show some state or disposition that has been

engaged in (Eagly & Chaiken, 1993). The assumption by researchers is that attitudes are

formed through a cognitive learning process where one gains information and then forms

beliefs. The information is gained through experiences with the object, such as the

Internet or a particular Website (Eagly, & Chaiken, 1993).

The most common way to measure attitude is through semantic differentials (Eagly

& Chaiken, 1993). During the development of this measure, researchers have found that

three factors are usually underlying the scales: evaluation, potency, and activity (Eagly &

Chaiken, 1993). The evaluative factor accounted for the most variability among scale

ratings analyzed and was identified to represent attitude. The bipolar-adjectives that load

in the evaluative dimension, like good/bad and pleasant/unpleasant, are thus used in

semantic differentials to measure attitudes (Eagly & Chaiken, 1993). Two researcher-

developed semantic differential scales were thus utilized. Attitude toward the treatment or

control version of the site to which subjects were exposed (Table 3-6) and the Internet in

general (Table 3-7) was tested through two sets of 11 semantic differential scales






64


(good/bad, pleasant/unpleasant, trustworthy/untrustworthy, effective/ineffective,

useful/not useful, and favorable/unfavorable) (Sicilia, Ruiz, & Munuera, 2005). These bi-

polar adjectives were placed at each end of a five-point scale. Three out of the eleven

attributes were reverse coded to decrease the influence of response layout (Dillman,

2000).

Table 3-6. Example of Scale Used to Measure Attitude toward the Treatment or Control
Version of the site to Which Subjects were Exposed
The information presented on this Website is


Good
Not credible
Biased
Difficult to
understand
Important
Not interactive
Easy to find
Not beneficial
Believable
Not trustworthy
Accurate


Bad
Credible
Unbiased
Easy to understand

Not important
Interactive
Hard to find
Beneficial
Unbelievable
Trustworthy
Inaccurate


Measure of attitude toward the Internet in general was measured for descriptive

purposes only.

Table 3-7. Example of Scale Used to Measure Attitude toward the Internet in General


I feel that many Websites on the Internet are
Good 1 2 3 4
Credible 1 2 3 4
Unbiased 1 2 3 4
Difficult to 1 2 3 4
understand
Not important 1 2 3 4
Not interactive 1 2 3 4
Easy to find 1 2 3 4
Beneficial 1 2 3 4
Believable 1 2 3 4
Trustworthy 1 2 3 4
Accurate 1 2 3 4


Bad
Not credible
Biased
Easy to understand

Important
Interactive
Hard to find
Not beneficial
Unbelievable
Not trustworthy
Inaccurate









An eight-item scale measuring the importance of the Internet in subjects' lives was

also utilized for descriptive purposes (Table 3-8). This index adapted by Ko (2001) from

Rubin (1985) and Conway and Rubin (1991) asked subjects to indicate level of

agreement with statements that discuss the importance of Internet in their lives. The index

has a reported internal reliability Cronbach's alpha of .86 (Ko, 2001).

Table 3-8. Example of Index Used to Measure the Importance of the Internet
Please rank your level of agreement with the following statements
I would rather surf the Internet than do 1 2 3 4 5
something else. Strongly Strongly
Disagree Agree
My knowledge increases as my 1 2 3 4 5
Internet usage increases Strongly Strongly
Disagree Agree
It would be very difficult for me to 1 2 3 4 5
survive without the Internet for several Strongly Strongly
days. Disagree Agree
Internet users are better educated 1 2 3 4 5
people. Strongly Strongly
Disagree Agree
The Internet opens doors that would 1 2 3 4 5
otherwise be closed. Strongly Strongly
Disagree Agree
Information online should be 1 2 3 4 5
engaging. Strongly Strongly
Disagree Agree
Information online should be 1 2 3 4 5
interactive. Strongly Strongly
Disagree Agree
Information online should be 1 2 3 4 5
entertaining. Strongly Strongly
Disagree Agree


Data Analysis

The data analysis for this study was completed using SPSS 12.0 for Windows PC.

Multiple analysis of variance (MANOVA) was utilized to allow for a more sophisticated

analysis of multiple independent and dependent variables (Graziano & Raulin, 2000).

MANOVA's allow for more complex examinations of the simultaneous relationships of

many variables, allowing researchers to create more sophisticated models to explain









social behaviors (Sweet & Grace-Martin, 2003). Several multivariate analyses of

variances were then used to compare means and interaction effects. Effect sizes of

univariate analyses of variances were calculated to describe the magnitude of treatment

effect (Kotrlik & Williams, 2003). Effect size reporting allows for judgment on the

magnitude of differences between groups, and allows for better comparison to previous

research results (Kotrlik & Williams, 2003).The Chohen's f, which estimates the

proportion of variance explained for the sample by the categorical variable was calculated

as follows: (Kotrlik& Williams, 2003)

o2 = SSbetween/SSTotal Cohen's f = Square root of (c 2/1 -c2)














CHAPTER 4
RESULTS

The purpose of this study was to examine the effects of problem-solving style, level

of Website interactivity, and Internet usage on an individual's attitude toward an

information-driven Extension Website and subjects' recall of the information presented

on that site. Based on a conceptual framework relating Kirton's Adaption-Innovation

(KAI) theory and the theory of Uses and Gratifications, research hypotheses were formed

with attitude and information recall as the dependent variables.

The instruments and experimental condition were administered to a sample

(N=314) of college undergraduates at a large Southeastern university. A total of 314

instruments were distributed in class. Those 314 participants were then sent either the

treatment or control condition and final instrument. A total of 305 participants returned

the final instrument for a 96.8% response. Cases were then removed based on the

following criteria:

1. The respondent had not fully completed all the instruments (n=l)

2. The respondent indicated on the KAI that nothing was easy or hard for them,
indicating their score was suspect (n=48).1


1 The KAI is a very sensitive instrument in which participants who respond that
everything is easy or hard for them or those who select down the middle of the scale must
be rejected as it is suspected that they are being reluctant to respond truthfully or are
trying to deliberately score in an "acceptable" way (Kirton, 1999, p. 19). It is noted by
KAI researchers (Tefft, 1994; Kirton, 1999) that younger populations will have lower
maturity levels, affecting the rejection rate of such a population. However, many
university students have enough work experience to understand the items without
problems (Kirton, 1999). A 10%-20% wastage rate can be expected under favorable
conditions with university students (Kirton, 1999).










This resulted in a final N of 256 participants.

Demographics

General demographics were calculated from the sample for gender, age, and

college rank (Table 4-1). There were 110 males (43.1%) and 145 female (56.9%)

respondents. The majority of respondents were 18-20 years old (56.3%), followed by

respondents 21-23 years old (37.9%), respondents 24-27 years old (5.1%), and

respondents 28 years or older (0.8%). There were 119 (46.9%) who reported being

college juniors, 66 (26%) sophomores, 61 (24%) seniors, and 8 (3.1%) freshmen.

Table 4-1. Number of Respondents by Age, Gender, and Class Rank
Characteristic N %
Age (n=256)
18-20 144 56.3
21-23 97 37.9
24-27 13 5.1
28+ 2 .8

Gender (n=255)
Male 110 43.1
Female 145 56.9

Rank(n=254)
Freshman 8 3.1
Sophomore 66 26.0
Junior 119 46.9
Senior 61 24.0


The majority (73.1%, n=187) of respondents indicated being enrolled in the

College of Agricultural and Life Sciences, followed by 9.8% (n=25) in the College of

Health and Human Performance, 4.7% (n=12) in the College of Business, 3.7% (n=10) in

the College of Liberal Arts and Sciences, 3.5% (n=9) in the College of Public Health and

Health Professions, 1.6% (n=4) in the College of Design and Construction Planning, .8%









(n=2) in the College of Pharmacy, and .4 % (n=l) in the College of Engineering and the

College of Medicine, respectively (Table 4-2).

Table 4-2. Number of Respondents by College (n=252)
College N %
College of Agricultural and Life Sciences 187 73.1
College of Health and Human Performance 25 9.8
College of Business 12 4.7
College of Liberal Arts and Sciences 10 3.7
College of Public Health and Health Professions 9 3.5
College of Design and Construction Planning 4 1.6
College of Pharmacy 2 .8
College of Engineering 1 .4
College of Medicine 1 .4
Undecided 1 .4


Media Selection and Internet Usage

When asked to indicate their preferred choices of media when seeking

information/news, 60.9% (n=156) indicated they preferred to use the Internet, while

26.6% (n=68) preferred television, 8.2% (n=21) preferred newspapers, 2.0% (n=5)

preferred magazines, 1.2% (n=3) preferred radio, and 1.2% (n=3) preferred information

from a book.

Participants were asked to describe their Internet and computer usage. The majority

(98.8%, n=253) indicated that they own a personal computer. High speed (55.3%, n=140)

and wireless access (37.5%, n=95) were the most indicated methods to access the Internet

at home, while at school the majority use a computer lab (49.8%, n=126) (Table 4-3).









Table 4.3. How Participants Access the Internet at Home and at Campus
Access n %
At Home (n=252)
High-speed 140 55.3
Wireless 95 37.1
Dial-up 16 6.3
Computer lab 1 .4
At Campus (n=253)
Computer lab 126 49.8
High-speed 65 25.7
Wireless 62 24.5


Respondents identified whether they personally had a Web log (blog), Facebook

page, MySpace page or a Website. The majority (89.5%, n=229) did not have their own

Website, and 89.1% (n=228) did not have a personal blog, while 85.2% (n=218) had a

page on Facebook, and 10.5% (n=27) had a page on MySpace. Of respondents, 82.4%

(n=210) indicated they had never created a Website.

Participants indicated their level of attitude toward the Internet in general through

semantic differentials. The Internet was seen to be moderately good, easy to understand,

important, easy to find, beneficial, believable and accurate (Table 4-4.). The grand mean

for general attitude toward the Internet was 3.2 (SD= .91) on a 1 to 5 scale (1 being

negative and 5 being positive).

Table 4-4. Level of General Attitude Toward the Internet
n M SD
Beneficial 256 3.7 1.0
Easy to Understand 255 3.6 1.0
Easy to Find 255 3.6 1.1
Good 256 3.5 .90
Interactive 254 3.3 .97
Important 254 3.1 1.0
Believable 256 3.1 .75
Accurate 256 3.1 .77
Credible 256 3.0 .80
Trustworthy 255 2.9 .76
Unbiased 255 2.4 .90









Extension Usage

Questions were asked on a 1 to 5 likert scale (strongly disagree to strongly agree) to

determine if subjects had experiences with Extension information. A mean of 1.96

(n=254) indicated that the majority of participants have not used Extension information in

the past, demonstrating that these participants were not heavy users of Extension services

in general. A mean of 2.11 (n=253) indicated that these participants have not visited the

University of Florida's Extension Website in the past. (See Table 4-5.)

Table 4-5. Mean Extension Experience of Study Participants*
n M SD
I have used Extension Information 254 1.96 1.21
I have looked at an Extension Website 253 2.11 1.40
*Based on a 1-5 Scale (1= strongly disagree to 5= strongly agree)

Message Relevance

When asked how important the information presented to them was, 160 (62.5%)

indicated that this information was moderately important to very important to them. A

total of 150 (58.6%) of the respondents indicated that they were moderately interested to

very interested in the information on car buying. The majority (50%, n=128) were

moderately knowledgeable on the topic of car buying. The majority (74.2%, n=190) have

not recently purchased a car, but 56.0% (n=141) have thought of purchasing recently. In

general, the majority of participants were interested in and moderately informed about

purchasing a car in the near future.

Manipulation Checks

Manipulation checks were conducted to evaluate the independent variables used in

the study. Based on the literature and findings from the pilot study, two versions of a

Website were developed containing the same information. Both versions contained facts

on car buying and an Extension identification image, but differed in the amount of









interactivity offered to the respondent. While car buying is not a traditional agriculture

message, the information presented was developed by agricultural economists and

presented through Extension. To assess the face and construct validity issues, participants

exposed to both versions were asked to identify along a five-point Likert scale if they

strongly agreed to strongly disagreed that the site version they viewed was interactive or

not. The means for both groups indicated that overall, on a Likert scale ranging between

1 and 5, they correctly identified the version they viewed as being either interactive or

non-interactive (Table 4-6). An independent sample t-test was run to test for the

significance of the condition manipulation with the version of the site serving as the

independent variable. Results showed a significant difference between the treatment and

control version of the sites at a .05 alpha level (Table 4-7).

Table 4-6. Means Table for Site Interactivity Identification*
Site is Interactive N Mean SD

Interactive Site 123 3.35 1.22
Non-interactive 132 1.96 1.11
Site
*Based on a 1-5 scale (1= strongly disagree to 5= strongly agree)



Table 4-7. Univariate Analysis of Variance for Significant Differences between Non-
Interactive and Interactive Version of the Sites.
df MSE F p
Non-interactive/interactive 1 1.37 86.55 .000


Problem-solving Inventory

The problem-solving instrument (KAI) used in this study, as described in Chapter

3, included 32 items. Respondents were asked to indicate the degree of ease or difficulty

they encountered in sustaining their adaptive and innovative behaviors over periods of









time by drawing an x where they fit in a five-point scale ranging from "very easy" to

"very hard." Scores ranged from 50 to 133 with a mean score of 92.6. As with many

other variables in psychology (Graziano & Raulin, 2000), the KAI is reported as a

continuous score, which are often preferred by statisticians for the ability to run "simpler"

calculations (Agresti & Finlay, 1997).

Grouped distributions are generally required when working with a continuous

variable such as KAI when reporting demographics (Graziano & Raulin, 2000). When

there are many possible scores the data reported in tabular form will be long and difficult

to read (Graziano & Raulin, 2000). Thus, a means split was conducted on the continuous

variable for descriptive reporting purposes only. The Theory of Adaption-Innovation is

deeply based on the idea that cognitive problem-solving style measured by the KAI is

based on a continuous scale, and should be treated that way in complex statistical

interpretations (Kirton, 1999). In order to formulate the means split, participants with

scores over 96 were considered to be more innovative, and those below 96 were deemed

more adaptive (Kirton, 2003). The means split resulted in 115 innovative participants

(M=105.34, SD= 7.64) and 141 adaptive participants (M=82.18, SD= 12.51) (Table 4-8).

Table 4-8. Means Table for Problem-solving Style Based on the KAI.
N (N%) Mean SD
Adaptive 141 (55.1%) 82.18 12.51
Innovative 115(44.9%) 105.34 7.64


Descriptive statistics were run on the categorizations of being more adaptive or

innovative based on the literature that females tend to be more adaptive than males

(Foxall & Haskins, 1986). This study supported the literature with more males (N=58)

who were more innovative than females (N=51) and more females (N=93) than males









(N=52) who were more adaptive in their problem-solving style (Table 4-9). The overall

mean KAI score for female participants in this study was 89.42 (SD=16.84) and for male

participants, 96.57 (SD=13.07).

Table 4-9. Means Table for the Effect of Gender on Problem-solving Style Based on the
KAI.
N Mean SD
Adaptive Male 52 85.94 8.89
Female 93 80.67 13.86
Innovative Male 58 106.10 7.78
Female 51 105.37 7.45


Internet Usage Constructs

As indicated in Chapter 3, a 13-item, researcher-developed construct was

developed to assess Internet usage. Respondents were asked to indicate on a five-point

Likert scale how many hours they spent online each day, how many sites they visited in

each stint online, if they have ever created a Website, and how often each week they

downloaded music, read a blog, instant-messaged, read Facebook/MySpace, watched

online videos, shopped online, used search engines or WebCT, and accessed news online.

Standard deviations for the scale ranged between .9 and 1.5, indicating a satisfactory

amount of variability in the scale. Based upon reliability analysis, all 13 items were

retained for an overall Cronbach's alpha of .73 (Table 4-10).

Table 4-10. Inter-item Consistency Statistics for the Internet Usage Construct (N=247)
Usage Item Mean* SD Corrected Item Alpha if item
total deleted
Correlation
How many hours a week do 2.1 .92 .53 .70
you spend on the Internet
How many sites do you visit 2.5 .91 .36 .71
on an average session online
Have you ever created a .20 .39 .28 .72
Website
Download music 2.4 1.33 .33 .71
Read a blog 1.9 1.13 .37 .70









Table 4-10 Continued. Inter-item Consistency Statistics for the Internet Usage Construct
(N=247)*
Usage Item Mean* SD Corrected Item Alpha if item
total deleted
Correlation
Instant message 3.5 1.52 .36 .71
Read Facebook or MySpace 3.9 1.37 .25 .73
Watch videos 2.3 1.20 .44 .70
Shop online 2.4 1.10 .33 .71
Shop/sell on Ebay 1.7 .94 .30 .71
Use a search engine 4.3 .91 .43 .70
Work on WebCT or other 4.3 .90 .17 .73
online course
How often do you use the 4.0 1.10 .48 .70
Internet to find
news/information online
*Five-point response scale where 1=very little to 5= very often.

After the data was summated, respondents were then categorized into a high or low

Internet usage based on a means split of 2.71. The median for the group was 2.70 and a

mode of 2.38. Based on the means split, 128 participants were considered to be lower in

their Internet usage and 119 were considered to be higher in their Internet usage (Table 4-

11).

Table 4-11. Means Table for Internet Usage.
n (n%) Mean SD
Low Internet Usage 128 (51.8%) 2.30 .30
High Internet Usage 119 (48.2%) 3.15 .32


Attitude Constructs

Based upon previous research, three attitudinal scales were developed to assess

general attitudes toward the Internet and attitudes toward the treatment or control version

of the site to which they were exposed. For the hypothesis testing, the scale measuring

attitudes toward the treatment or control version of the site to which they were exposed

was utilized. As discussed in Chapter 3, the most often used way to test attitude is

through semantic differentials (Eagly & Chaiken, 1993). Two 11-item semantic-









differential scales were used with bi-polar adjectives placed at the end of five-point

scales.

The scale measuring attitude toward the Internet in general was utilized for

descriptive demographic reporting only and showed standard deviations from .8 to 1.1,

indicating a satisfactory amount of variability. The coefficient alpha reliability for the

index was a=. 70 (Table 4-12). The summated mean for the overall scale was 3.2 (SD=

.91). Indicating a moderately positive attitude toward the Internet in general.

Table 4-12. Inter-item Consistency Statistics for the Attitude Toward the Internet in
General (N=249)
Usage Item Mean* SD Corrected Item total Alpha if item
Correlation deleted
Good 3.5 .88 .40 .67
Credible 3.0 .80 .43 .67
Unbiased 2.4 .90 .20 .70
Easy to 3.6 1.00 .22 .70
Understand
Important 3.1 1.00 .45 .66
Interactive 3.3 .96 .20 .70
Easy to Find 3.6 1.10 .21 .70
Beneficial 3.6 1.00 .47 .66
Believable 3.1 .76 .42 .67
Trustworthy 2.8 .77 .43 .67
Accurate 3.0 .77 .50 .66
*Five-point response scale where 1=very little to 5= very often.

Attitude toward the treatment or control version of the site to which they were

exposed was measured after exposure to the version of the site on an 11-item semantic

differential scale. Based upon the reliability analysis, 10 of the items were retained. The

standard deviations for the scale ranged from .78 to 1.1 (Table 4-13). The coefficient

alpha reliability score for the index was a=.80. The summated mean for the overall scale

was 3.88 (SD=.94).









Table 4-13. Inter-item Consistency Statistics for the Attitude Toward the Treatment or
Control Version of the Site to Which They were Exposed (N=237)
Usage Item Mean* SD Corrected Item total Alpha if item
Correlation deleted
Good 4.0 1.0 .49 .74
Credible 3.7 .94 .45 .75
Unbiased 3.7 .93 .35 .76
Easy to 4.2 .97 .41 .75
Understand
Important 3.8 1.0 .47 .74
Interactive 3.9 1.1 .32 .77
Easy to Find 3.9 .96 .54 .74
Beneficial 4.1 .88 .48 .74
Believable 3.8 .78 .56 .74
Trustworthy 3.7 .82 .56 .74
*Five-point response scale where negative to 5= positive.

The last scale which was also utilized for demographic description only was an

eight-item index adapted from Rubin (1985) and Conway and Rubin (1991), where

participants indicated their level of agreement along a five-point scale of the importance

of the Internet in their lives. A reported internal reliability of .86 was reported in previous

research (Ko, 2001). The standard deviations for the scale in this study ranged from .77 to

1.51. The coefficient alpha reliability score was ca=.72 (Table 4-14). A summated mean

for the overall scale was 3.27 (SD= 1.02).









Table 4-14. Inter-item Consistency Statistics for the Importance of the Internet in
Subjects' Lives (N=237)
Usage Item Mean* SD Corrected Alpha if item
Item total deleted
Correlation
I would rather surf the 2.63 1.02 .35 .70
Internet than do something
else.
My knowledge increases as 3.02 1.05 .41 .69
my Internet usage increases.
It would be very difficult for 3.10 1.51 .42 .70
me to survive without the
Internet for several days.
Internet users are better 2.54 1.15 .45 .68
educated people.
The Internet opens doors that 4.03 .95 .46 .68
would otherwise be closed.
Information online should be 3.70 .77 .50 .68
engaging.
Information online should be 3.54 .84 .40 .69
interactive.
Information online should be 3.63 .86 .37 .70
entertaining.
*Five-point response scale where strongly disagree to 5= strongly agree.


Information Recall

For the purpose of hypothesis testing, data on unaided recall was utilized. For the

unaided recall portion of the study, participants were asked to list all of the information

they could recall from the treatment or control version of the site to which they were

exposed. The resulting qualitative data was content analyzed and all true statements were

scored as a +1 while untrue statements were coded as a -1 (Davis et al, 2005). Scores

ranged from -2 to 16 (n=255). Two individuals indicated items that were not presented on

the site, so their statements were coded negatively. A grand mean of 4.27 (SD=2.88) was

calculated (Table 4-15).









Table 4-15. Descriptive Report for Unaided Recall (N=255)
N Minimum Maximum Mean SD
Information Recall 255 -2.00 16.00 4.27 2.88


Hypotheses Tests

Several hypotheses were made based on the independent and intervening effects of

Internet usage, problem-solving style, and site interactivity on subjects' attitude and

information recall. An overall means table (Table 4-16) provides insight into the average

attitudes toward the treatment/control version of the site split by low/high level of

Internet usage, adaptive/innovative problem-solving, and the version of the site given.

Problem-solving was grouped by adaptive and innovative rather than showing the

variable as continuous to help with readability of the statistics (Graziano & Raulin,

2000).

Table 4-16. Means for Attitude Toward Treatment/Control split by Low/High Level of
Internet Usage, Adaptive/Innovative Problem-solving Style, and Experimental
Condition Presented (With Cell Sizes)
Experimental More Adaptive More Innovative
Condition
Low Internet High Internet Low Internet High Internet
Usage Usage Usage Usage
Interactive 3.45 3.59 3.57 3.39
(36) (35) (22) (16)
Non-Interactive 3.57 3.54 3.44 3.58
(32) (30) (29) (30)
Total 3.50 3.57 3.50 3.51
(68) (65) (51) (46)


An overall means table (Table 4-17) shows the average information recall split by

low/high level of Internet usage, adaptive/innovative problem-solving, and the condition

presented with.









Table 4-17. Means for Information Recall split by Low/High Level of Internet Usage,
Adaptive/Innovative Problem-solving Style, and Experimental Condition
Presented (With Cell Sizes)
Experimental More Adaptive More Innovative
Condition
Low Internet High Internet Low Internet High Internet
Usage Usage Usage Usage
Interactive 4.74 4.33 4.30 3.78
(38) (36) (23) (18)
Non-Interactive 4.15 4.12 4.18 4.58
(34) (33) (33) (31)
Total 4.46 4.23 4.23 4.29
(72) (69) (56) (49)


H1: Unaided information recall and level of attitude toward an information-driven
Extension Website will differ significantly as a function of site interactivity,
problem-solving style, and Level of Internet usage.

It was predicted that problem-solving style, level of Internet usage, and site

interactivity will affect the attitude toward an information-driven Extension Website and

the information recalled from that site. To test this, a multivariate analysis of variance

(MANOVA) was run (Table 4-18). MANOVAs offer a more sophisticated analysis of the

variables allowing for the exploration of multiple independent and dependent variables

(Graziano & Raulin, 2000). Results showed a partial support for this hypothesis. No

significant three-way interaction was found between problem-solving style, level of

Internet usage, and site interactivity on attitude (F=.67, p=.80) or information recall

(F=1.50, p=. 13). Results also indicated no two-way interactions between problem-solving

style and site interactivity on attitude (F=.81, p=.75), problem-solving style and Internet

usage on attitude (F=.65, p=.92), or site interactivity and Internet usage on attitude

(F=.26, p=.61). However, significant two-way interactions were found between problem-

solving style and site interactivity on information recall (F=1.60, p=.05), problem-solving









style and Internet usage on information recall (F=1.84, p=.01), and site interactivity and

Internet usage on information recall (F=9.53, p=.00).

Results indicated no main effects for problem-solving style on attitude (F=.64,

p=.97), site interactivity on attitude (F=.05, p=.83) and information recall (F=.40, p= .53),

Internet usage on attitude (F=.02, p=.89) and information recall (F=.21, p= .65).

However, a significant main effect for problem-solving style on information recall was

found (F=2.12, p=.00).

Table 4-18. MANOVA Results for Problem-solving Style, Site Interactivity, and Level of
Internet Usage on Attitude and Information Recall (N=229)
Source Df F P
Problem-solving Style (PS) Recall 61 2.12 .00*
Attitude 61 .64 .97
Level of Internet Usage (IU) Recall 1 .21 .65
Attitude 1 .02 .89
Site Interactivity (SI) Recall 1 .40 .53
Attitude 1 .05 .83
PS x SI Recall 31 1.60 .05*
Attitude 31 .81 .75
PS x IU Recall 36 1.85 .01*
Attitude 36 .65 .92
SI xIU Recall 1 9.53 .00*
Attitude 1 .26 .61
PS x SI x IU Recall 14 1.50 .13
Attitude 14 .67 .80
Error Recall 80 (5.55)
Attitude 80 (.31)
Note. Values enclosed in parentheses represent mean square errors.


Across the whole design, significant two-way interactions between problem-

solving style and site interactivity were found for information recall, indicating the

degree to which one is more adaptive/innovative and the level of site interactivity

affected how much information was recalled. A significant two-way interaction was also

found for problem-solving style and Internet usage for information recall, indicating that









the degree to which one is more adaptive/innovative and their level of Internet usage may

have an effect. Means tables further demonstrates the relationship between problem-

solving style and site interactivity (Table 4-19) and problem-solving style and Internet

usage (Table 4-20).

Table 4-19. Means for Level of Problem-solving Style and Site Interactivity on
Information Recall Overall
Site Interactivity Problem-solving Mean N Std. Deviation
Non-interactive Adaptive 4.13 67 2.97
Innovative 4.38 64 2.60
Interactive Adaptive 4.47 78 2.77
Innovative 4.00 46 3.31


Table 4-20. Means for Problem-solving Style and Internet Usage on Information Recall
Overall
Usage Problem-solving Mean N Std. Deviation
Low Adaptive 4.46 72 3.08
Innovative 4.23 56 2.96
High Adaptive 4.23 69 2.65
Innovative 4.29 49 2.97


The significant two-way interaction between site interactivity and level of Internet

usage on information recall suggests that information recall may differ based on level of

Internet usage and site interactivity. A means table further demonstrates the relationship

between site interactivity and Internet usage (Table 4-21).

Table 4-21. Means for Site Interactivity and Internet Usage on Information Recall
Site Interactivity Usage Mean N Std. Deviation
Non-interactive Low 4.16 67 2.74
High 4.34 64 2.85
Interactive Low 4.57 61 3.30
High 4.15 54 2.70


The significant main effect shows that differences may lie in the level of problem-

solving style. More adaptive individuals had a mean recall of 4.32 (SD=2.86) and more









innovative individuals had a mean recall of 4.22 (SD=2.91) (Table 4-22), indicating that

individuals who were more adaptive recalled information better overall.

Table 4-22 Means for Problem-solving Style on Information Recall
Problem-solving N Mean Std. Deviation
Adaptive 141 4.32 2.86
Innovative 105 4.22 2.91


For Subjects who Received the Interactive Site

H2: Attitude toward an information-driven Extension Website will differ
significantly as a function of Internet usage and problem-solving style.

It was expected that when holding site interactivity constant to include only those

subjects who were exposed to the interactive version of the site, attitudes would differ as

a function of Internet usage and problem-solving style. Results show no support for the

hypothesis. The ANOVA results (Table 4-23) show no two-way interaction between

problem-solving and Internet usage (F=.68, p=.84). No main effects were found for

problem-solving (F=.62, p=.94) or Internet usage (F=.00, p=.97) on attitude toward an

information-driven Extension Website.

Table 4-23. ANOVA Results for Those Viewing the Interactive Version, Problem-
solving Style, and Internet Usage on Attitude Toward and Information-Driven
Extension Website (N= 109)
Source df F P
Internet Usage (IU) 1 .00 .97
Problem-solving Style (PS) 47 .62 .94
PS x IU 23 .68 .84
Error 37 (.34)
Note. Values enclosed in parentheses represent mean square errors.


H3: Unaided information recall will differ as a function of Internet usage and
problem-solving style.

It was expected that when holding site interactivity constant to include only those

subjects who were exposed to the interactive version of the site information recall would









differ as a function of Internet usage and problem-solving style. Results show support for

the hypothesis. The ANOVA results (Table 4-24) show a significant two-way interaction

between problem-solving and Internet usage (F=2.04, p=.02, cohen's f= .59). Significant

main effects were found for problem-solving (F=2.19, p=.00) and Internet usage (F=9.77,

p=.00) on information recall.

Table 4-24. ANOVA Results for Those Viewing the Interactive Version, Problem-
solving Style, and Internet Usage on Information Recall (N=115)
Source Df F P
Internet Usage (IU) 1 9.77 .00*
Problem-solving Style (PS) 49 2.19 .00*
PS x IU 23 2.03 .02*
Error 41 (5.25)
Note. Values enclosed in parentheses represent mean square errors.


For those receiving the interactive site, information recall differed significantly

based on problem-solving style and Internet usage. A means table sheds light on the

interaction between problem-solving and Internet usage (Table 4-25).

Table 4-25. Means for Problem-solving Style and Internet Usage on Information Recall
for Individuals Viewing the Interactive Version of the Site
Usage Problem-solving Mean N Std. Deviation
Low Adaptive 4.74 38 3.02
Innovative 4.30 23 3.78
High Adaptive 4.33 36 2.53
Innovative 3.78 18 3.06


A means table further describes the main effects of Internet usage (Table 4-26) and

problem-solving styles (Table 4-27). Based on the means table, subjects lower in Internet

usage had slightly higher information recall than those higher in Internet usage.









Table 4-26. Means for Internet Usage Main Effects on Information Recall for Those
Viewing the Interactive Version of the Site
Source Level Mean N Std. Deviation
IU High 4.15 54 2.70
Low 4.57 61 3.30
Those who are more adaptive had higher information recall than subjects who were

more innovative.

Table 4-27. Means for Problem-solving Style Main Effects on Information Recall for
Those Viewing the Interactive Version of the Site
Source Level Mean N Std. Deviation
PS Adaptive 4.47 78 2.78
Innovative 4.00 46 3.30


For Subjects Who Received the Non-Interactive Site

H4: Attitude toward an information-driven Extension Website will differ
significantly as a function of Internet usage and problem-solving style.

It was expected that when holding site interactivity constant to include only those

subjects who were exposed to the non-interactive version of the site, attitude would differ

as a function of their level of Internet usage and problem-solving style. Results show no

support for the hypothesis. The ANOVA results (Table 4-28) show no two-way

interaction between problem-solving and Internet usage (F=.50, p=.97). No main effects

were found for problem-solving (F=.93, p=.60) or Internet usage (F=.40, p=.53) on

attitude toward an information-driven Extension Website.

Table 4-28. ANOVA Results for Those Viewing the Non-Interactive Version of the Site,
Problem-solving Style, and Internet Usage on Attitude Toward and
Information-Driven Extension Website (N=121)
Source Df F P
Internet Usage (IU) 1 .40 .53
Problem-solving Style (PS) 48 .93 .60
PS x IU 27 .50 .97
Error 44 (.29)
Note. Values enclosed in parentheses represent mean square errors.









H5: Unaided information recall will differ significantly as a function of Internet
usage and problem-solving style.

It was expected that when holding site interactivity constant to include only those

subjects who were exposed to the non-interactive version of the site, unaided information

recall would differ as a function of their level of Internet usage and problem-solving

style. Results show partial support for the hypothesis. The ANOVA results (Table 4-29)

show a significant two-way interaction between problem-solving and Internet usage

(F=1.69, p=.05, Cohen's f= .53). Significant main effects were found for problem-

solving (F=1.65, p=.04), while no significant main effects were found for Internet usage

(F=1.58, p=.22) on information recall.

Table 4-29. ANOVA Results for Those Viewing the Non-Interactive Version, Problem-
solving Style, and Internet Usage on Information Recall (N=115).
Source Df F P
Internet Usage (IU) 1 1.58 .22
Problem-solving Style (PS) 49 1.65 .04*
PS x IU 31 1.69 .05*
Error 49 (5.50)
Note. Values enclosed in parentheses represent mean square errors.


Results indicate that for those receiving the non-interactive site, information recall

differs based on problem-solving style and Internet usage. A means table sheds light on

the interaction between problem-solving style and Internet usage (Table 4-30). For those

viewing the non-interactive site, innovators are higher in their information recall

regardless of their level of Internet usage.




Full Text

PAGE 1

EFFECT OF COGNITIVE PROBLEM-SO LVING STYLE, INTERNET USAGE, AND LEVEL OF INTERACTIVITY ON ATTITU DES TOWARD AND RECALL OF WEBBASED INFORMATION By EMILY B. RHOADES A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Emily B. Rhoades

PAGE 3

This document is dedicated to my husband Aaron. Thank you.

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my committee members for all of their time, effort, and advice through this process. Th eir insight has made my res earch stronger. I thank Mindy McAdams for her wise advice on the Internet and new technologies that she brought to this study. I thank Brian Myers for his methodol ogical and statistical help and advice on surviving the process of a di ssertation. I thank Ricky Telg for his friendship, mentorship, and time in helping me through every stage of this study. Lastly, I thank my advisor and chair, Tracy Irani, for meeting with me endl ess times when she had many other things going on. Her advice and guidance have not only strengthened this study, but have developed me as a researcher and academic. I am thankful to my friends and family for all of their suppo rt through the ups and downs of the last few years. I thank my fr iends back home, Marie and Erin, for reminding me that life is richer than th e things I was experiencing in school. It was nice to know I could always call for a reality check. I want to thank my 310 cohorts for their advice, friendship, and encouragement. I could not have done it without Shannon, Courtney, and Wendy, thanks for putting up with me on the bad days and celebrating with me on the good days. Thanks to everyone for the trips to get ice cream and the weekend nights at the Jones. I am indebted to my Florida fa mily. I look forward to many future research conferences with everyone.

PAGE 5

v I want to thank my parents for always being on the other end of the phone after a bad day. I thank them for encour aging me to stay strong and develop myself as a person. Their guidance and love has taken me furt her in life than I ever expected. Finally, I thank my husband for standing by me through the last few years of school. I appreciate him moving to Florida a nd putting our life on hold. His dedication to me and our future is amazing. His friendship and love has encourag ed me through all of this, and I hope someday I can repay him. Thank you.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES..........................................................................................................xii ABSTRACT ....................................................................................................................xi ii CHAPTER 1 INTRODUCTION........................................................................................................1 Introduction to the Study..............................................................................................1 Communication and Agri cultural Science....................................................................2 Extensions Role in Co mmunicating Agriculture.........................................................5 Internet as an Information Source.................................................................................8 Website Design and Structure....................................................................................11 Uses and Gratifications...............................................................................................13 Previous Usage with Media.................................................................................14 Attitude................................................................................................................15 Information Recall...............................................................................................16 Problem Solving.........................................................................................................16 Purpose of the Study...................................................................................................19 Key Terms..................................................................................................................20 Organization...............................................................................................................20 2 LITRATURE REVIEW..............................................................................................22 Overview.....................................................................................................................22 Interactivity and Linearity..........................................................................................22 Interactivity..........................................................................................................23 Linearity.......................................................................................................28 Adaption/Innovation Theory......................................................................................29 Problem Solving..................................................................................................29 Kirtons Theory...................................................................................................30 Information Richness...........................................................................................36 Uses and Gratifications Theory..................................................................................37 Uses and Gratifications and the Internet.............................................................39

PAGE 7

vii Previous Experience and Expectations................................................................41 Information Recall......................................................................................................42 Attitude....................................................................................................................... 44 Conceptual Framework...............................................................................................46 Research Questions.....................................................................................................47 3 METHODOLOGY.....................................................................................................48 Overview.....................................................................................................................48 Hypotheses..................................................................................................................48 For Subjects Who Receive the Interactive Site:..................................................49 For Subjects Who Receive the Non-Interactive Site:..........................................49 Research Design.........................................................................................................49 Subjects....................................................................................................................... 52 Pilot Study..................................................................................................................53 Procedure....................................................................................................................54 Instrumentation...........................................................................................................56 Independent Variables.........................................................................................57 Treatment.....................................................................................................57 Problem-solving style...................................................................................59 Internet Usage..............................................................................................60 Dependent Variables...........................................................................................62 Information Recall........................................................................................62 Attitude.........................................................................................................63 Data Analysis..............................................................................................................65 4 RESULTS...................................................................................................................67 Demographics.............................................................................................................68 Media Selection and Internet Usage....................................................................69 Extension Usage..................................................................................................71 Message Relevance.............................................................................................71 Manipulation Checks..................................................................................................71 Problem-solving Inventory.........................................................................................72 Internet Usage Constructs...........................................................................................74 Attitude Constructs.....................................................................................................75 Information Recall......................................................................................................78 Hypotheses Tests........................................................................................................79 For Subjects who Received the Interactive Site..................................................83 For Subjects Who Received th e Non-Interactive Site.........................................85 5 DISCUSSION.............................................................................................................88 Overview.....................................................................................................................88 Key Findings...............................................................................................................90 Implications of the Study............................................................................................94 Problem-Solving Style.........................................................................................94

PAGE 8

viii Internet Usage......................................................................................................98 Limitations................................................................................................................100 Recommendations for Theory and Practice..............................................................103 Recommendations for Practitioners..................................................................103 Future Research.................................................................................................109 Conclusions...............................................................................................................114 APPENDIX A INSTRUMENTS......................................................................................................116 B EXPERIMENTAL CONDITION............................................................................125 LIST OF REFERENCES.................................................................................................131 BIOGRAPHICAL SKETCH...........................................................................................146

PAGE 9

ix LIST OF TABLES Table page 3-1. Independent Sample T-test for Signi ficant Differences Between Courses and Version of the Site Based on Age or Gender............................................................53 3-2. Means Table for Website In teractivity Identification.................................................54 3-3. Independent Sample T-test for Si gnificant Differences Between Less and Interactive Version of the sites.................................................................................54 3-4. Example of KAI Instrument Items.............................................................................60 3-5. Example of Internet Usage Items...............................................................................61 3-6. Example of Scale Used to Measure At titude toward the Treatment or Control Version of the site to Which Subjects were Exposed...............................................64 3-7. Example of Scale Used to Measure Att itude toward the Internet in General.............64 3-8. Example of Index Used to Measur e the Importance of the Internet...........................65 4-1. Number of Respondents by Age, Gender, and Class Rank........................................68 4-2. Number of Respondents by College (n=252).............................................................69 4.3. How Participants Access the Internet at Home and at Campus..................................70 4-4. Level of General Attitude Toward the Internet..........................................................70 4-5. Mean Extension Experi ence of Study Participants.....................................................71 4-6. Means Table for Site Interactivity Identification.......................................................72 4-7. Univariate Analysis of Variance for Significant Diffe rences between NonInteractive and Interactiv e Version of the Sites........................................................72 4-8. Means Table for Problem-sol ving Style Based on the KAI.......................................73 4-9. Means Table for the Effect of Ge nder on Problem-solving Style Based on the KAI...........................................................................................................................7 4

PAGE 10

x 4-10. Inter-item Consistency Statistics for the Internet Usage Construct (N=247)...........74 4-10 Continued. Inter-it em Consistency Statistics for the Internet Usage Construct (N=247)..................................................................................................................75 4-11. Means Table for Internet Usage...............................................................................75 4-12. Inter-item Consistency St atistics for the Attitude Towa rd the Internet in General (N=249)..................................................................................................................76 4-13. Inter-item Consistency Statistics for the Attitude Toward the Treatment or Control Version of the Site to Which They were Exposed (N=237).....................77 4-14. Inter-item Consistency St atistics for the Importance of the Internet in Subjects Lives (N=237)........................................................................................................78 4-15. Descriptive Report for Unaided Recall (N=255)......................................................79 4-16. Means for Attitude Toward Treatm ent/Control split by Low/High Level of Internet Usage, Adaptive/Innovative Probl em-solving Style, and Experimental Condition Presented (With Cell Sizes)..................................................................79 4-17. Means for Information Recall split by Low/High Level of Internet Usage, Adaptive/Innovative Problem-solving Style, and Experimental Condition Presented (With Cell Sizes)...................................................................................80 4-18. MANOVA Results for Problem -solving Style, Site Interactivity, and Level of Internet Usage on Attitude a nd Information Recall (N=229)................................81 4-19. Means for Level of Problem-solving Styl e and Site Interac tivity on Information Recall Overall........................................................................................................82 4-20. Means for Problem-solving Style and Internet Usage on Information Recall Overall....................................................................................................................82 4-21. Means for Site Interactivity and Internet Usage on Information Recall...................82 4-22 Means for Problem-solving Style on Information Recall..........................................83 4-23. ANOVA Results for Those Viewing the Interactive Versi on, Problem-solving Style, and Internet Usage on Attitude Toward and Information-Driven Extension Website (N=109)...................................................................................83 4-24. ANOVA Results for Those Viewing the Interactive Versi on, Problem-solving Style, and Internet Usage on Information Recall (N=115)....................................84 4-25. Means for Problem-solving Style and In ternet Usage on Information Recall for Individuals Viewing the Inter active Version of the Site........................................84

PAGE 11

xi 4-26. Means for Internet Usage Main Effects on Information Recall for Those Viewing the Interactive Version of the Site...........................................................85 4-27. Means for Problem-solving Style Main Effects on Information Recall for Those Viewing the Interactive Version of the Site...........................................................85 4-28. ANOVA Results for Those Viewing the Non-Interactive Version of the Site, Problem-solving Style, and Internet Usage on Attitude Toward and Information-Driven Extension Website (N=121)..................................................85 4-29. ANOVA Results for Those Viewing th e Non-Interactive Version, Problemsolving Style, and Internet Usag e on Information Recall (N=115).......................86 4-30. Means for Problem-solving Style and In ternet Usage on Information Recall for Individuals Viewing the Non-Inte ractive Version of the Site...............................87 4-31. Means Problem-solving Main Effects on Information Recall for Those Viewing the Non-Interactive Version of the Site.................................................................87

PAGE 12

xii LIST OF FIGURES Figure page 2-1. Conceptual framework for this study.........................................................................47 3-1. A screen capture of the consent informati on and instructions sent to participants....55 3-2. A screen capture of the non-interactiv e control page sent to participants..................57 3-3. A screen capture of the interactive versi on of the site page sent to participants........58

PAGE 13

xiii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECT OF COGNITIVE PROBLEM-SO LVING STYLE, INTERNET USAGE, AND LEVEL OF INTERACTIVITY ON ATTITU DES TOWARD AND RECALL OF WEBBASED INFORMATION By Emily B. Rhoades August 2006 Chair: Tracy Irani Major Department: Agricultur al Education and Communication This study examined the effects of pr oblem-solving style, level of Website interactivity, and Internet usag e on an individuals attitude toward an information-driven Extension Website and subjects recall of th e information presented on that site. This study is based on a conceptual framework re lating Kirtons Adapti on Innovation Theory and Uses and Gratifications. Successful problem-solving is in demand in the area of agriculture. As Extension services and communicators move to designing in formation online, it is cruicial that this information be presented in a form that will be usable, valuable, appropriate, and easy to recall. By understanding how problem-solving styles affe ct users perceptions of Websites, with respect to such attributes as attitude and recall information, Extension, agricultural communicators, a nd commodity groups who are ut ilizing the Internet to

PAGE 14

xiv reach audiences will be better able to develop communications processes that match audience needs in order to inform them, e ducate them, and effect productive change. This study shows that problem-solving styles, coupled with an individuals Internet usage have an affect on information recall. While researchers continue to debate if interactivity affects at titude and recall of information, these findings show no individual effects of interactivity on attitude and in formation recall when presenting informationdriven content to a young adult population. Howe ver, it was found that for interactive or non-interactive versions of an informationdriven site, informa tion recall could vary based on problem-solving style and level of Internet usage. There are populations such as innovative problem-solvers who retain information better from the non-interactiv e versions of online Extens ion information. The more adaptive individuals will actua lly do better with less stru cture and ambiguity when working online successfully. It is also noted th at for low users of the Internet, the novelty of interactivity attracts and keeps the interest of users to increase their retention of information, as supported by the literature These findings encourage designers of information-driven sites to take inventory of how they are presenting their information to specific audiences.

PAGE 15

1 CHAPTER 1 INTRODUCTION Introduction to the Study A cow that was cleared of having ma d cow disease last fall by the U.S. Department of Agriculture was in fact inf ected with the brain-wasting disease, the department announced Friday, making it the second confirmed case of the disease in this country The Chicago Tribune reported on June 25, 2005 (U.S. Confirms Mad Cow, 2005). Communicating information a bout science has always been an important endeavor, but perhaps never as crucial as it is today. In an age which the technology to create the science seems to be in a race to outp ace the technology used to communicate it, communicators must be set to address the chan ging needs of audiences. Today, the public is faced with many information choices, presented through a multitude of different communication channels. One such channel that has emerged as an important tool for those seeking science information is the Internet (Morris & Ogan, 1996). Science-based information has the potential to create situations that save lives (Henroid, Ellis, & Huss, 2004). After completi on of formal education, most people will only be exposed to science through chance encounters with news reporting (Treise & Weigold, 2002) and occasional informal educat ion. Much of the information presented through mass media informs audiences about breakthroughs in science, food safety, medicine, and technology, all topi cs that have the potential to greatly influence lifestyles. While science information is usually t hought of as being related to biology, biotechnology, food science, or horticulture, science topics covered by organizations like Extension also include social sciences and economic research. One such topic covered by

PAGE 16

2 Extension science information includes budge ting and purchasing of large items like automobiles (University of Iowa, 1998). The Internet can offer new challenges to those communicating science-based information to audiences (Henroid, Ellis, & Huss, 2004). Through the power to integrate other forms of media on the Web, communicat ors are now able to present complex, science-based information in new ways, su ch as interactive video, animations, and graphics, which create memorable learni ng experiences for viewers. Finding and retrieving information is considered to be more convenient by online users who are able to find information on many topics in a shor t amount of time (Henroid, Ellis, &Huss, 2004). Along with these new opportu nities, however, come many challenges. Research is just beginning to examine the answers as to how user characteristics relate to the processing of design and display of information online. Given the importance of effectively conveying scientific information, researchers must examine the best methods available to help people successfully discove r and interpret the information they find online. Communication and Agricultural Science Todays world is science-driven, and for the benefits of scientific advancements to be dispersed, publics need to be able to interpret and understand that information (Shortland & Gregory, 1991). A societys understa nding of this information is important not only for the well-being of its citizens, but also for th e continued support of these endeavors. Educated publics should be ab le to choose between conflicting reports on information concerning scientific adva ncements (Treise & Weigold, 2002). By effectively communicating science to audiences, favorable attitudes that are created

PAGE 17

3 toward science and science funding will allow fo r a clearer understanding of the benefits that science adds to society (Treise & Weigold, 2002). Scientific information tends to be comple x, detailed accounts of new advancements or findings. Research has shown that scie nce communicators often frame the news by only reporting on the breakthroughs (Gunter Kinderlerer, & Beyleveld, 1999). For example, researchers, looking at the opinions of scientists and journa lists, found that both groups agreed media coverage of biotechnol ogy information was questionable, taking into account the complexity of the subj ect (Gunter, Kinderler er, & Beyleveld, 1999). Poor reporting of this information has been a concern of scientists and researchers alike (Gunter, Kinderlerer, & Beylevel d, 1999; Treise & Weigold, 2002). As a subset of the scientific community, ag ricultural science is an important aspect of science with respect to Americas ec onomy and environment (Ruth, Telg, Irani, & Locke, 2004). Many current scientific issues th at have been extensively reported on, such as mad cow disease, biotechnology, and animal cloning, are all deep ly embedded within agriculture. Developments in agriculture over th e last few years have created many opportunities, as well as challenges, to rese archers and communicators. Agricultural and science information affects everyone on an ever yday basis, whether they are aware of it or not (Saunders, Akers, Haygood, & Lawv er, 2003; Lundy, Ruth, Telg, & Irani, 2005). It is vital that this information is percei ved accurately by the ge neral public, due to the significant impact of agriculture on society a nd public health (Terry & Lawver, 1995). For generations, agriculture has been intertwined with greater human society by serving as a support and underpinning (Paw lick, 2001). However, as important as the

PAGE 18

4 information is, many argue that agriculture science is minimally covered through the media (Pawlick, 2001). The changes in agri culture and its impact on the American economy make the need for communicati ng agriculture crucial for creating an agriculturally literate publi c (Lundy, Ruth, Telg, & Irani, 2005, p. 6). Frick, Birkenholz, and Machtmes (1995) also contend that ev ery person should possess a minimum amount of knowledge and understanding of the scientific i ndustry that provides food for human survival. The decreases in the farming and ranch populations have made this an even more vital need, as members of the general public, far removed from the rural setting, may no longer have accurate perceptions of agriculture (Saunders, Akers, Haygood, & Lawver, 2003). Consumers need to be literate on agricultural issues in order to respond aptly when issues that deal with the inters ection of food and scienc e occur. For this to take place, however, these issues must be communicated effectively. The agricultural industry, the federal a nd state-mandated land grant institutions, the farm press, and academic journals continue to disseminate information on agricultural issues. Land grant institutions in particular, through their Cooperative Extension Services and public information specialists, are charged with the task of technology transfer and disseminating public-value information to th eir clientele and members of the general public (Seevers, Graham, Gamon, & Conklin, 1997). Until the advent of the Internet, Extension communicators were forced to use traditional mass media as their primary communications channel. However, traditi onal media gatekeepers, often focusing on news value as a function of their audiences demographics and occ upations, have been viewed by some as not always highly valui ng agricultural news, and information with

PAGE 19

5 respect to determining what is on the pub lic, and consequently, the media agenda (Cartmell, Dyer, Birkenholz, & Sitton, 2004). Extensions Role in Communicating Agriculture As discussed earlier, one such organizati on charged with comm unicating science to audiences is the U.S. Cooperative Extensi on Service. Extension links education and research resources of the land-grant unive rsity, the United Stat es Department of Agriculture (USDA), and county administrative units to provide scientific information to non-formal educational audien ces (Seevers et al., 1997). The Smith-Lever Acts of 1862, 1890, and 1990 formally integrated these resources to provide instruction and information in agriculture and home economics and related subjects to those not currently taking cour ses in the Land Grant colleges. The land grant College system was established through the Mo rrill Acts in 1862 and 1890 to provide at least one college in each state that would ha ve a leading objective to provide learning opportunities related to ag riculture and the mechanic arts to the industrial class (Seevers, Graham, Gamon, & Conklin, 1997). Cooperative Extension has changed its fo cus over the years to include broader initiatives and issues by serving as problem so lvers in issues of the environment and other social and economic changes going on in communities (Seevers et al., 1997). While Extension programs were historically seen as rural programs, they now cover almost every aspect of peoples lives They have also begun moving to reach greater audiences, including rural and non-rural audiences of all ages th rough new technological and communication outlets (Seevers et al., 1997). Researchers over the last several ye ars have been providing evidence of Extensions need to embrace Internet technol ogies to reach audiences (Howell & Harbon,

PAGE 20

6 2004). The broader audiences and community-b ased needs being addressed by Extension in the 20th century caused researcher s and Extension professionals to look into new and more cost-effective methods of informa tion dissemination (Bull, Cote, Warner, & McKinnie, 2004; Wood-Turley & Tucker, 2002) The need to provide these diverse audiences with timely, pertinent information that allows them to maintain a working dialogue with these audiences has caused many Extension communicators to develop communication via the Internet (Siegrist, La barge, & Prochaska, 1998). Several studies on Extension have encouraged the continue d movement online. Kaslon, Lodl, and Greve (2005) studied the effectiveness of online leaders training for 4H volunteers, and found that the Internet is a good method to reach these audiences with training and continued education. Lippert, Plank, and Radhakrishna (2000) looked at the effectiveness of regional Internet Extension in-service traini ng to reach agents. The researchers found that it was not only successful in knowledge acquisiti on, but was also seen as just as effective as face-to-face administration by participan ts. Dunn, Thomas, Green, and Mick (2006) found that an interactive online multimedi a Extension product can increase knowledge and influence behaviors on nutrition with hi gh school students. They recommended that multimedia was a good way for Extension to educate young people on health-related topics. In a consumer focus group study focusing on the value of Extension services, Irani, Ruth, Telg, and Lundy (2005) recommended th at Extension adopt the communication technology used by their target audiences. Th e researchers also noted that for the participants they asse ssed, that technology was the Web. Th e ability of the Internet to provide cost efficient information that can reach large audiences has been described as a

PAGE 21

7 valuable tool for Extension and its clientel e, especially with y ounger audiences (Jackson, Hopper & Clatterbuck, 2004). A study of landowners in 2004 indicated age was a significant factor as to whether the public wanted information on watershed conservation issues via the Internet or through writte n communication. Younger landowners showed a higher preference for computer-based info rmation (Howell & Harbon, 2004), while the majority of respondents still preferred traditional written communication. Howell and Harbon (2004) concluded that while curre nt trends may still prefer traditional communication methods, Extension must move toward targeting th ese younger audiences who will be the landowners of the future. Bull and colleagues, in 2004, called for Ex tension to pay specific attention to underserved audiences who have typically not been original stakeholders in the program. Young adults ranging in age from 18 to 24 ar e among one of the groups not traditionally serviced through Extension programs that ar e traditionally aimed at youth, pre-college, and adult homeowners (Seevers, et al, 1997). While research has shown that younger adults are some of the lowest users of Extens ion, they are interested in Extension services such as community development programs (Warner, Christenson, Dillman, & Salant, 1996). Audiences of young adults are also the fu ture users of Extension programs as they graduate and become involved in communities and purchase homes. Nationally, Extension has attempted to answer this call to move online by introducing the e-Xtension initiative, led by the Extension Comm ittee on Organization and Policy (MSState, 2005). The goal of this program is to implement a national Webbased information and education network fo r all Extension clientele (MSState, 2005). This modern marketplace will facilitate engagement with audiences in new subject areas

PAGE 22

8 in a manner that is accessible and timely. User s of the system will be prompted to provide information that allows them to receive personalized assistance for the information contained in the 3,000-plus counties in the U.S ., and yet still be connected to their state and local Extension organi zations (MSState, 2005). By focusing on answering users questions and problems, plans are for e-Xtension to provide information in various interact ive formats, including frequently asked questions, fact sheets, chat sessions, discussion boards streaming video, Web-based conferencing, and educational modules (e -Xtension, 2005). By providing convenient, quick-access to information, the goal is that us ers will be able to solve problems and find information to improve their daily lives, t hus supporting the missi on of Extension. The foundation for this initiative was set in place and in September 2005, communities of practice, or topic areas that will be focused upon, were developed. These included parenting, horticulture, disa ster education, financial s ecurity, local economics and entrepreneurship, wildlife management, fire ants, and equine re sources (e-Xtension, 2005). Internet as an Information Source One reason organizations like Extension are moving online is because of the dramatic increase in Internet usage. Niel sen/NetRatings reported that 135.82 million Americans were active Inte rnet users in 2004 (ClickZ, 2005). As of February 2003, Americans spent an average of 25.5 hours pe r month using the Internet (CyberAtlas, 2003). Recent studies have shown that one major use of the Internet by these populations has been for news information. The Web has been found to be the third-most important source of news following radio and newspa pers (Chan & Leung, 2005). It was noted in

PAGE 23

9 2003 that 40% of adults in the United States us e the Web to get news, weather, and sports information (Lieb, 2005). A survey done by the Pew Research Center found that onethird (and almost half of those under 30) of respondents now receive news information online at least once a week (Bogart, 2000). Conw ay (2001) reported that more than four out of every 10 respondents to his study we re using computers to find out what was happening in the world. In a survey of 400 Mi dwest university studen ts, it was found that 47.8% use the Internet frequently for referen ce or research materials (Bressers & Bergen, 2000). Stempel, Hargrove, and Bernt (2000) found a decline in the use of local and network television and newspapers while ther e was a huge gain in Internet use by the general public. Chan and Leung (2005), on th e other hand, found that heavy users of newspapers and radio tended to spend a longe r amount of time online reading news than light users. Communicators are attempting to reach th eir changing audiences by offering online news and information. Garrison (2001) found, th at as of 1999, almost 90 % of U.S. daily newspapers were actively using new online tech nologies to reach new markets. It can be assumed this number only continues to grow as more users are logging on. In a national study done by the Pew Internet & American Life Project, 80% of all Americans said they would expect to find information online about health and news (Horrigan & Rainie, 2002). One in five Americ ans revealed that they rely heavily on the Internet to find information. Of those who go to the Internet for government and health information, the majority was white females with children under th e age of 18 (Horrigan & Rainie, 2002). According to a recent surv ey of Internet users, 49% are college graduates, 38% have family incomes over $75,0 00, and their main transaction online is to

PAGE 24

10 gather information or for entertainment (Fallows, 2004). Men are more likely than women to use the Internet for information, while women use it more to communicate. In 2005 it was noted that 68% of males and 66% of females were Web users, and 80% of males and 86% of females ages 18-19 we re using the Internet (Burns, 2005). College students represent the largest population of Internet users (Eastin, 2001) making them an important subset to study. An overwhelming 96% of all 18-29 year-old users find the Internet a good way to get inform ation, compared to 91% of all older users (Fallows, 2004). And with more than 15 milli on college students in the United States who represent a $9.2 billion market for consum er goods (Ness, Gorton, & Kuznesof, 2002) this audience is one to pay close attention to. Fallows (2004) found in a recent survey of Internet users that 49% are college gradua tes, making future graduates an important audience segment to study. When the Internet began growing as a communication outlet, Morris and Ogan (1996) called for scholars to rethink definitions and categ ories of communication and mass media in terms of the new technology. Webster and Lin (2002) also found that the Internet is a viable communi cation outlet that should be looked at by researchers. Scholars have answered this call with resear ch in this area; however, many questions remain unanswered. Dibean and Garrison (2001, p. 88) concl uded, development of the technology of the Internet and the Web itself may become the most significant change in world communication in a half-century or longer. Based on the growing influence of this new medium, researchers and communicators need to understand how users best process

PAGE 25

11 information presented online. The Internet has created the opport unity for new methods of news delivery by combining components of print and broadcast media (Berry, 2001). Although use of the Internet has been studied in the context of news and information dissemination in general, limite d research has been done in the area of agricultural communications, in particular, in terms of provi ding information online that is preferred and is recalled by audiences. Da vis and colleagues (2005) found that adults studied recalled agricultural communication information better when presented to them in a print, video, or radio news release over electronic text. Based on a thorough review of the literature, however, few studies have assess ed how the level of interactivity of online, Web-based Extension communicat ion efforts and ones psycho metric traits, such as problem-solving style, affects the inform ation recall and att itudes of users. Website Design and Structure Many researchers have begun to study the In ternet from a visual communication, or graphical and structural, perspe ctive in order to analyze how Websites are using design to reach general audiences, as well as how certain components on these site s aid in recall of information and perceived preference (C ho, 2003; Bogart, 2000; Diao & Sundar, 2004; Lang, Borse, Wise, & David, 2002). Abraham (2001) argued that online communication, by its very nature, is a presentation that is driven by visuals and visual communicators. Esrock and Leichty (1999) call for communi cators to think of their users and to develop sites that are not only efficient in te rms of technology, but also visually pleasing. In a medium that allows for displaying gra phics and multimedia, it is easy to provide information on pages that are pleasing to vi ew and easy to naviga te (Henkia, 1990). Few people want to dig through confusing pages of information, and designing a site that is

PAGE 26

12 easily navigated can complement communi cations by keeping viewers coming back (Henkia, 1990). Users are looking for simplicity and usabil ity as they enter sites (Nielsen, 2000). Swenson, Constantinides, and Gurak (2002) described a need to use logical design choices and define the audience members of the site in order to better reach them. Websites are a visual media, in which factor s such as layout, de sign, and graphics can either add credibility to an organization and aid in information intake, or hinder individuals ability to pro cess information (Amant, 2005). Recent studies in the area of Internet me dia have focused on the effect of non-linear based information (Lowrey, 2002; Dimitrova Connolly-Ahern, Williams, Kaid, & Reid, 2003; Tremayne, 2004). In one such study, Lo wrey (2002) found that non-linearity did not affect perceived credibil ity or knowledge acquisition. Tewksbury and Althaus (2000) compared the non-linear reality of online news to print editions of the same newspaper. Findings showed that online readers were le ss likely to recall events of national, international, and political importance than those reading traditional print-based publications. In agriculture, the research has been less ex tensive in this area; the look and feel of sites hosted by communication organizations has been under-researched (Williams & Woods, 2002). In a research synthesis of th e Journal of Applied Communications from 1992-2001, Williams and Woods (2002) noted that a large portion of published research analyzed the readership trends of agricu ltural communication outlets, but has ignored their design or Web presences. Researchers have targeted agricultural communicators through practitioner-oriented articles in the Journal of Applied Communications, the

PAGE 27

13 leading journal in this applied field, written to help them design effective Websites for their audiences, but little rese arch has been presented on de signing effective sites using features such as interactivity (Emery, 1999; Kelleher, Henley, Gennarelli, & Hon, 1997; Melgares, 2005). Uses and Gratifications Researchers have described communicati on behavior as be ing goal-directed, purposive, and motivated (Rubin, 1994). The Web has been described by several scholars as being a medium that requires active us ers (Bouwman & Wijngaert, 2002; Kaye & Johnson, 2002; Papacharissi & Rubin, 2000). Users of the Web are challenged with finding the information that brought them to that site. Uses and gratifications, as a theoretical perspective, describes users psychological and social environmental needs, their need s and motivations to communicate, the media they choose, their attitudes toward that media, the alternatives to that media, their communication behavior, and their outcomes are all important elements in the communication process (Rubin, 1994). By initiati ng the selection of and use of a specific media vehicle, users are actively seeking out in formation in order to fulfill a need (Rubin, 1994). The most salient use of the Web seen by researchers has been the informationseeking function (Papacharissi & Rubin, 2000). Th is theory can play a role in assessing Web use as a function of sociability, prur ience, curiosity, and information-seeking (Ruggiero, 2000). Uses and gratifications disc usses the cognitive processe s that take place between the complexities of needs felt by individua ls, such as solving problems or making decisions and how users gratify those n eeds through media (Blumler, 1979). Graber (1984) argues that those who are drawn to me dia information to receive gratification are

PAGE 28

14 more likely to be able to lear n that information. In order to discover what aids people in recall when using a specific media, it is imperative that research ers keep in mind the motives that bring users to that media and how they perceive it. Recall has shown to differ as a result of media consumed (Dav is et al, 2005; Tewks bury & Althaus, 2000; Eveland & Dunwoody, 2001), and by understanding what brings users to media, it can aid in encouraging users to go to the right kind of media that will help them recall information. Uses and gratifications scholars have examined many motives for using the Internet, and found in genera l that the Internet tended to satisfy entertainment, information, and interaction needs (Papach arissi &Rubin, 2000; Ko, 2002). Kaye and Johnson (1998) found that 40% of respondents used the Web primarily for information and education research. Kaye and Johnson (2002 ) concluded that respondents were active users who sought out specific information via s earches and interacted with others through chat rooms and listserves. Previous Uses and Gratifications studies found, political attitudes were strongly linked to measures of information seeking and surveillance (Kaye & Johnson, 2002). Thus, through the advent of the Internet, Uses and Gratifications researchers have begun retesting items found to be salient wi th respect to other media (Ruggiero, 2000). Ruggiero argued that the advent of the In ternet would only incr ease the theoretical potency of Uses and Gratifications by allo wing researchers to explore the theoretical linkages with respect to this new communications medium. Previous Usage with Media According to the Uses and Gratificati ons theory, previous experience and gratifications met can give us ers a respective image of that medium and what they can

PAGE 29

15 expect from it (Katz, Blumler, & Gurev itch, 1974). Peled and Katz (1974) found in a study of media during wartime and crisis that people came to a specific medium with expectations of what that medium will be a nd what it will gratify for them. While a large percentage of usage and gratifications rese arch has explained previous usage in the context of traditional media channels, much new research is being conducted with the Internet (Ruggiero, 2000; Baran & Davis, 2003). Many recent studies have utilized Uses and Gratifications as part of the theoretical framework when studying the Internet because of the interactivity, demassificati on, and asynchronicity it allows that other media outlets do not (Baran & Davis, 2003). With 135-plus million users actively using the Internet in 2004 (ClickZ.com, 2005), it could be the case that they are all coming to the medium with preconceive d ideas of the qualities of that media source. Attitude Within the Uses and Gratifications paradigm, attitudes are formed based on the experience and the gratifications met as a f unction of choosing a particular medium to serve a specific need or information-seeking function. Attitude has been shown to be an important predictor of usage and implem entation of technology and continued use (Rodgers & Chen, 2002). Attitude research ha s been done extensively in the area of advertising and attitude toward advertis ements and their effectiveness (Chen, 1999; Rodgers & Chen, 2002; Sinclair & Iran i, 2005; MacKenzie & Lutz, 1989), and researchers have begun employing these same tasks to look at Internet sites and advertising (Rodgers & Chen, 2002; Chen, 1999). Rodgers and Chen (2002) reported adoption of the Internet by advertising agen cies is affected by poor attitudes toward and lack of experience with Internet advertis ing. Chen (1999) developed a scale to provide researchers the ability to measure the attitude s of users of Websites to help indicate the

PAGE 30

16 value of such sites. For the purposes of this study attitude was defined as: A psychological tendency that is expressed by evaluating a particular entity with some degree of favor or disfavor (Eagly & Chaiken, 1993, p. 1). Information Recall As discussed earlier, recall can be incr eased as a function of the enhanced gratification from a medium. Several studies have looked at the Internet and information recall. Lowery (2002) reported th at linearity of sites had an effect on degree of perceived control over media, but it did not affect the degree of perceived cr edibility or recall of knowledge. DHaenens, Jankowski, and Heuvelm an (2004) reiterated this, saying that news category, gender, and interest played mo re of a role in recall than whether the information was given via online or in prin t. For the purpose of this study information recall is defined as the learning process a nd free recall of information, not just aided identification. Problem Solving Problem solving has been described as a survival skill in todays technological world (Wu, Custer, & Dyrenfurth, 1996). Kirton (2003) described problem-solving as the means by which life survives and manages th e constant change pr esented through ones environment. Problem solving has also been defined as the tendency to respond in a certain way while addressing problems (Wu, et al, 1996, p. 55). This linear process of evaluation begins with users recognizing a specific problem, defining it, having the ability to comprehend and develop it, test hypotheses and gather da ta about it, revise those hypotheses and retest, and then form a conclusion on the problem (Hedges, 1991). Part of the process of problem-solving incl udes gathering informa tion through channels such as the Internet. As one begins this pr oblem-solving process, it is important to note

PAGE 31

17 that individuals are limited by the way they are built in terms of intelligence, but they also have no instincts to he lp or hinder them in this (Kirton, 2003, p. 33). However, individuals are intelligent within different st yles that allow them to problem solve given the opportunity or motivation (K irton, 2003). As they work through this linear process of problem-solving they are driven by our indivi dual problem-solving style (Kirton, 2003). Style is something that is uni que to individuals as a psychom etric quality while process is the structure in which all indivi duals go about solving problems. A great deal of research has been done on the cognitive and decision-making processes that bring people to specific media. One aspect of this approach has been to look at the concepts of information richness a nd equivicality. Information richness theory claims that individuals choose media they percei ve to be most efficient in helping them to complete a communication task or problem (Kelleher, 2001). People cognitively choose which medium will help them as they work to solve problems or gratify other needs. A rich medium can be described as containi ng more face-to-face type interactions, while leaner media are seen as being formal and num erical, such as telephones versus fliers or fact sheets (Trevino, Daft, & Lengel, 1990). Those in a high equivocality condition will tend to choose a richer Website more often th an leaner communications media, such as a pamphlet or a lean Website (Irani & Kelle her, 1997). People will choose rich media based on equivocality, which can be assumed to be related to the innate differences in which people solve problems. Trevino, Daft and Lengel (1990) further described equivocality by comparing an equivocal message to a Rorschach ink-blotmultiple users may read it differently than others depending on their unique backgrounds and perspectives. This theory rela tes to many of the current comp onents of interactivity that

PAGE 32

18 are seen on Websites. As a Website has more in teractivity, it will be richer and have more equivicality, as opposed to a leaner site with more ambiguity. Another way researchers look at the phenom enon of how individuals choose media to solve problems is through cognitive styles. Cognitive activity refers to the degree of mental activation invoked in paying atte ntion to a medium (Gunter, 2000, p. 164). Cognitive style has been described by researchers as also referring to a persons consistent pattern of processing informati on and organizing it into a system of thought which influences behavior (Foxall & Hask ins, 1986, p. 65) when they are working to solve problems or make decisions. The cogni tive framework allows us to explore the levels of involvement with media (Gunter, 2000). Much research has been conducted in this area examining the visual representation of broadcast news; however, some researcher s have begun to use these same methods to analyze online information (Fox et al., 2004; Lang, Borse, Wise, & David, 2002; Diao & Saunder, 2004; Gunter, 2000). G unter (2000) stated that to fully understand the use of media, one must assess the nature of the expos ure in terms of the cognitive effort put into the processing of the content. Once this is known, one can assess and measure the media influence on awareness and knowledge gain ed by the audience (Gunter, 2000). Researchers have tested specific medi a outputs to determine the amount of information recalled and the cognitive effort pl aced into viewing. Consideration has been given to how information is cognitively proc essed and recalled based on the effects of how that information is presented (Gunter 2000). Sicilia, Ruiz, and Munuera (2005) looked at the effects of Intern et non-linearity with respect to need for cognition, or an individuals tendency to engage in and en joy cognitive activity (Petty & Cacioppo, 1979).

PAGE 33

19 Sicilia and colleagues (2005) found that indi viduals who were exposed to interactive Websites processed information more thoroughly than those exposed to non-interactive sites. For example, people looking at a site that was non-linear and required interaction would process the information presented mo re thoroughly than those presented with a linear site. One researcher who has focused on probl em-solving styles based on cognitive processing is Michael J. Kirton. Kirton (2003) states that people can be placed on a continuous scale, with individua ls who are adaptive in thei r problem-solving style at one end and those who are innovative in their prob lem-solving style at the other end. This theory defines and measures the thinking st yle that influences ones decision-making process (Kirton, 1999). According to Kirton, the adaptor will solve pr oblems within his or her existing perceptual frame of reference, while innovators will change those frameworks and do things differently as they seek solutions outside of the context of the given problem (Goldsmith, 1984; Foxall & Haskins, 1986). Researchers measure the adaptive-innovative dimension of cognitive style with the Kirton Adaption-Innovation Inventory (KAI) (Foxall & Bhate, 1993). Purpose of the Study Problem solving has been discussed in va rious contexts; however, limited research has been conducted on how problem-solving st yle aids in the decision-making process with respect to processing media that may provide information, especially information that can be used to solve problems from th e standpoint of recall of information and attitude toward the value and appropriaten ess of the information and media utilized. The study postulates problem-solving styl e, as conceptualized by Kirtons KAI inventory, will influence the way information seekers go about fulfilling their information

PAGE 34

20 needs based on previous knowledge/use of me dia, and in turn will influence their attitudes toward the type of media that users prefer and gain knowledge from. One such area where problem-solving is in demand is the area of agriculture. As Extension services and comm unicators move to designing in formation online to reach audiences and inform them about topics that wi ll help solve problems, it is important that this information be presented in a form that will be usable, valuable, appropriate, and easy to recall. By understanding how problem-solving styl es affect users perceptions of Websites, with respect to such attributes as attitude and information recall, Extension, agricultural communicators, a nd commodity groups who are ut ilizing the Internet to reach audiences will be better able to develop communications processes that match audience needs in order to inform them, educate them, and affect productive change. If users vary due to their problem-solving style, it may be the case that communicators need to provide information that appeal specifically to these diverse styles. Key Terms Information-Driven Website : The objective of informa tion-driven Websites is to provide the user with desired information. The objective of these sites is to guide the user to the desired content pages. Navigation pages support the user in his search (Stolz, Viermetz, Skubacz, & Neuneier, 2005, p. 1). Interactivity: A process involving users, medi a, and messages, with an emphasis on how messages relate to one anothe r (Sundar, Kalyanaraman,, & Brown, 2003, p. 34). Organization Chapter 1 has introduced the problem to be examined, as well as the purpose of the study. Chapter 2 will discuss in detail the rele vant literature and theoretical framework to be used in the study. These will include the Adoption Innovation model, Uses and

PAGE 35

21 Gratifications, Website interactivity and usage, attitude, perceptions of the Internet, and information recall. Chapter 3 will outline th e research design and methods of the study, including hypotheses, independe nt and dependent variables, description of participants, instruments and reliability of scales, procedur es, and statistical analysis used. Chapter 4 gives the results of data analysis performe d to test stated hypotheses. Chapter 5 will describe the limitations of the study, the results, and provide conclusions and recommendations.

PAGE 36

22 CHAPTER 2 LITRATURE REVIEW Overview The purpose of this investigation was to examine the effects of individuals problem-solving style, level of Website inte ractivity, and level of Internet usage on subjects attitude and amount of information r ecall with respect to an information-driven Extension Website. By understanding how problem-solving styles affect users perceptions of a Website with respect to such attributes as attitude and information recall, Extension, agricultural communicators, and commodity groups w ho are utilizing the Internet to reach audiences will be better able to utilize communications processes to inform audiences, educate them, and affect productive change. The following literature review explores the various components of cognitive problem-solving style as it rela tes to the study, as well as wh at drives individuals to choose specific media in certain situations, wh at design factors can be manipulated in an online environment by the organizations posting the information, and what influences an individuals attitudes and le vel of information recall. Interactivity and Linearity A significant amount of research in the advertising, communication, and marketing literature has focused on Website aesthetics, usability, and design. Within these domains, many researchers have addressed the ques tion as to how to make Websites more appealing to audiences. Resnick and Montania (2003) used the effects of semiotics, the study of signs and visuals, in Web design f eatures to explore the expectations of

PAGE 37

23 performance criteria in a purchase situation. The authors found that some design features have a strong effect on expectations of c onsumers. Cho (2003) found in a study of online banner advertisements that peripheral cues su ch as advertisement size and animation had an effect on those likely to click-through when they had high involvement with a product, while Thompson and Wassmuth (1999) cautioned that the use of tr ick banners on sites might lead to possible negative reactions. By ensuring good usability, design, a nd easy navigation through a site, communicators and Web developers can attend to their audiences. As users feel more comfortable with a site and are successful in gratifying their needs, they will like that site more, and they will return to the site agai n for information (Spool et al, 1999). While a site may be designed to effectively reach audi ences with information, it is still up to the user as to how they use that information. As Nielson (2000) descri bed, users have been shown to scan information online as opposed to engaging in deep c ognitive processing. It has also been noted by authors in the field of Website design that no two users will have the same experience with a site (Krug, 2000). Interactivity Interactivity is something that has be en defined by various scholars in many disciplines to mean different things. Heet er (1989) entered th e discussion early on commenting on how mass media has changed wi th the onslaught of new media like the Web. He discussed how the idea of mass has changed as those who look at a Website may not see the same thing as someone else looking at the same site, due to what he describes as interactivity and hypertext. Heet er (1989) set up several components of what he determined was a multidimensional concept, the first being complexity of choices available to the user which is defined by how many opportu nities there are for users to

PAGE 38

24 make decisions and to be in control and activ e. Along that same line is what he described as the effort exerted by the user He also described that interactivity takes into account information relative to users needs and the feasibility to have interpersonal communication. Heeter said that the measuring the users interactions while online also comes into play. And lastly, the ease of adding information is also to be considered. Beyond those steps Heeter went on to desc ribe how this new component of media changes the role of sender and receiver by making them interchangeable. Downes and McMillan (2000) added to this definition of interactivity. While they too expressed the importance of the changi ng role of sender a nd receiver and the importance of choice and effort exerted, they we nt further with this thought. They looked at interactivity in terms of its impacts, its nature, and its participants. The authors described how the new ideal of interactivity was having a large impact on how media and business operated, and would change industrie s. Downes and McMillan discussed the nature of the message included in interactivity in terms of its time needed and whether it was synchronous or asynchronous, how c onducive it was to two-way communication between sender and receiver, and in terms of its place and whether it created a sense of place. They considered the amount of control th e participant had (which is similar to that described by Heeter in 1989), the responsiveness to the needs of the user, and the goal of the interactivity. The authors stated that inte ractivity could be seen as being a continuum for these components, such that, when sense of place or sense of control increased, so did amount of interactivity. Research er stated that even the simp lest Website contains some interactivity as the user has control of wh at they see through the use of hypertext and links.

PAGE 39

25 Jensen (1998) also added to the discussi on as he described how interactivity should be looked at in several ways including through psychology, informatics, and mass communication. He described that interactivit y could be a criterion in which something must be present, or a form of communication offering users th e ability to get closer to interpersonal communication. He also describe d it as a technology. While this scholar also discussed the amount of control and e ffort exerted by the users he agreed with Downes and McMillan (2000) that interactivity should be looked at as a continuum or as dimensions. Jensen described four dimensi ons that range from one-way communication to n-way communication, which is similar to interpersonal communication and is continuous. He described four types of co mmunication in interactivity, which includes registration (two-way with feedback based on what the user inputs) continuous (two way communication), consultation (pre-defined c ontent a user seeks out with a feedback loop), and allocution (one-way communication with no feedback). He stated that for something to be interactive, it must be described by how much effort the user gives (Jensen, 1998). McMillan and Hwang (2002) went on to look at interactivity more closely, focusing specifically on four components. Si milar to Jensen, McMillan and Hwang said that one should look at interactivity th rough mass communication and consider the message and the four types of communicati on (registration, conti nuous, consultation, and allocution). As researchers look at this multidimensional concept, McMillan and Hwang argued that they should consider organizat ional communication a nd theorists Gruning and Grunings (1989) four-part model. McMill an and Hwang (2002) stated that one must look at the direction and components of the co mmunication as describe d by the theorists.

PAGE 40

26 He went on to say that researchers must al so look at it in terms of interpersonal communication and how people look at and us e technology, and lastly in terms of media components and what technology is actually involved. Liu (2003) attempted to develop a scale to measure the interac tivity of Websites. Three studies conducted resulted in three correlated dimensions of interactivity, including active control, two-way communication, and sync hronicity which comprised a scale to be applied to marketing and scholarly research. Researchers have taken the previously disc ussed descriptions of interactivity to look further into how interactivity on Website s affects users. Sundar, Kalyanaraman, and Brown (2003) examined interactivity as a cont ingency view. They defined interactivity as a process involving users, media, and me ssages, with an emphasis on how messages relate to one another (Sundar et al., 2003, p. 35). They conducted an experiment in which they broke interactivity into three levels: low interactive which contained no links, medium interactivity which was a single laye r of related links, and high interactive which had two hierarchal layers of links. Results showed that participan ts viewing the three conditions did not differ in their ability to recall and recognize content from a site (Sundar et al., 2003). They did find that the le vel of interactivity of the site had an influence on the impression of the political candidate featured on the site. They concluded that users perceptions of inter activity were positively associated with the number of hyperlinks present on the site and the linking actio ns initiated (Sundar et al., 2003). Bezjian-Avery, Calder, and Iacobucci (1998) also concl uded that interactivity may not always help users. They found in a study of advertising that in certain conditions, interactivity actually interrupted persuasion as user s could move right through the

PAGE 41

27 interactive media without at tending to the advertising message. Their study concluded that the linear traditional advertisements yi elded more positive or in some cases similar results with respect to deci sions to purchase. Liu (2003) found that different consumers may want different levels of interactivity in different situations. Individuals who are low interaction-ready consumers will prefer lower levels of interactivity than those individuals who are high in teraction-ready (Liu, 2003). Sicilia, Ruiz, and Munuera (2005) exam ined the effect of interactivity on information processing and favorability toward a product and found cont radicting results. In this study, interactivity was conceptu alized as a Website containing six pages connected through hyperlinks and email, and a non-interactive site set up as a print advertisement with the message on one page an d no hyperlinks (Sicilia et al., 2005). They concluded that individuals viewing the inte ractive site processed information more thoroughly than those viewing the non-interac tive site. It was found that motivation to process the information increased under the inte ractive condition. Teo, Oh, Liu, and Wei (2003) investigated th e effect of intera ctivity on attitudes toward commercial Websites. Results showed that interactivity on a site had a positive effect on a users perceived satisfaction and at titude toward a site. They also noted that interactivity levels significantly influenced th e sites effectiveness in helping users in a decision-making process (Teo et al., 2003). Chung and Zhao (2004) echoed these results in an experiment testing th ree levels of site hyperlinki ng and interactivity. They found that perceived interactivity had a positive influence on attitude toward the advertisement and memory of the information on the advertisement. Wu (1999) also looked at interactivity in advertisements online and c oncluded that perceive d interactivity had a

PAGE 42

28 positive effect on attitude. Wu defined inte ractivity in terms of responsiveness and navigability. Chen and Yen (2004) discovered th at interactivity on a site is related to viewers perceptions of the quality of site design. They suggested that successful sites should include interactive feat ures that add playfulness, connectedness, and reciprocal communication. Linearity Based on some definitions of interactiv ity, levels of site hyperlinking can be considered interactive, thus making it important to look at the literature on linearity. Several researchers have researched linear ity and hyperlinking in Websites (Tremayne, 2004; Lowrey, 2002; Dimitrova, Connolly-A hern, Williams, Kaid, & Reid, 2003; Eveland & Dunwoody, 2001; Berger, 2001; Ma ssey, 2004). Dimitrova and colleagues (2003) found in a study of newspaper Website s focusing on the coverage of Timothy McVeighs execution that online papers may use hyperlinks as a gatekeeping device. Other researchers have looked at linearity and its effects on users. Eveland, Cortese, Park, and Dunwoody (2004) found through an experiment with college students and adults that the non-linear structure of the Web has bot h strengths and wea knesses in terms of learning when compared to print media or more linear Websites. Researchers also found that linear site designs encouraged more f actual learning while non-linear sites increased knowledge structure density (Eveland et al, 2004). In contrast, Lowrey (2002) showed that user recall had no significant difference based on viewing a linea r or non-linear site. Lowrey also explained that linear structur e had an effect on the degree of perceived control over the media, but did not affect the perceived credibility. Sicilia, Ruiz, and Munuera (2005) studie d how consumers processed information and their experiences with in teractive and non-inte ractive Websites (whi ch they described

PAGE 43

29 as non-linear and linear, resp ectively). Experimental finding s showed that interactive (non-linear) sites lead to more processing of the information and more favorability toward the site. Berger (2001) discovered, in an experiment looking at hypertext, that a significant positive correlation existed between hypertext comfort and user satisfaction. However, the researcher also reported th at hypertext did not co rrelate with users information recall or accuracy in recall. Eveland and Dunwo ody (2001) found similar results in that no significa nt differences across linear and non-linear Websites were shown in terms of cued recall. Adaption/Innovation Theory One way to look at interactiv ity and linearity that coul d explain the differences in findings could be focusing on how individuals look at the informa tion based on specific psychometrics like cognitive style. Cognitive style as a theoretical construct is used to describe and explain an indi viduals processing of information when solving problems (Foxall & Bhate, 1993). Problem Solving Problem solving has been described as a skill needed for continued existence in todays technological world (Wu, Custer, & Dyrenfurth, 1996), the means by which life survives and manages the constant change presented in everyday life (Kirton, 2003). Problem solving can be seen as the inclina tion to respond in a cer tain way when faced with a problem (Wu et al, 1996). A problem has been defined by Goldsmith and Matherly (1986) as a situation where stan dard or customary procedures can not cope with the task due to unfamiliar elements that encroach. He dges (1991) described it as a linear process that begins with users recogni zing a specific problem, defini ng it, having the ability to comprehend and develop it, test hypotheses and gather data about it, revise those

PAGE 44

30 hypotheses and retest, and then form a c onclusion on the problem The data-gathering portion of the process is deeply embedded into the usage of communication tools. As one begins this process, individuals are limited by the way they are built, in terms of intelligence (Kirton, 2003). Wu and colleagues (1996) found evidence that differences between technological problem-sol ving and personal problem-solving style may exist. Personal problem-solving was de fined as problems dealing with depression, conflict, and life decisions (Wu et al, 1996) The researchers claimed that problemsolving style is an important difference betw een individual college students that must be looked at in terms of stude nts study of technology. Kirtons Theory Researchers have describe d that all people are bound by their makeup as to how they define and solve the problems with which they are faced (Kirton, 1999). This cognitive problem-solving style refers to the characteristic manner of how an individual will behave over situations and time. This c onsistent pattern of processing information influences behavior by or ganizing it into a system (Foxall & Haskins, 1986). One way to measure this cognitive styl e is through the Kirton Adaption-Innovation Inventory (KAI) (Foxall & Bhate, 1993). This inventory requires respondents to assess the degree of ease or difficulty they enc ounter in sustaining adaptive or innovative behaviors over periods of time. Responses are computed into overall scores ranging from 32 to 160 (Foxall & Haskins, 1986). Respondent s who score below the 96 midpoint are considered adaptors while those above 96 are innovators (Foxall & Haskins, 1986). Kirton (2003) states that people can be placed on a continuous scale between being adaptive and innovative in their problem-solving style. Cognitive style is a trait that can be expected to be stable over time and acr oss situations (Kirto n, 2003) People, however,

PAGE 45

31 may adapt their style through coping behaviors wh en they find themselves in a particular situation (Kirton, 1999). For example, one may be seen as being more adaptive at work and more innovative with friends. Those who ar e further apart on the KAI scale are more likely to have problems working together du e to these differences; however, when this happens, people begin to employ coping behaviors so that they are able to avoid these problems in some situations (Kirton, 2003). The adoption-innovation theory suggests th at adaptors and in novators voluntary styles of cognition differ in three respects: ru le conformity, efficiency, and preference for sufficiency versus proliferation of solu tions to problems (Fox all & Hackett, 1992). Adaptors will solve problems within their ex isting perceptual frame of reference, while innovators will change those frameworks and do things differently as they seek solutions outside of the context of th e given problem (Goldsmith, 1984; Foxall & Haskins, 1986). The KAI can be broken into three sub-s cales: sufficiency of originality (SO), efficiency (E), and rule/group conformity (R) (Kirton, 1999). When looking at the three subsets of KAI, adaptors and innovators can be described more in-depth. On the originality scale, adaptors tend to presen t only a few solutions to problems while the more innovative person may propose many, possi bly impractical, solu tions (Bagozzi & Foxall, 1995). The more adaptive individual w ill prefer to progress incrementally toward a goal and an innovator will avoid attention to detail when dealing with efficiency. Lastly, when comparing the rule governance subset, more innovative individuals ignore rules or invent their own rules as they go while more adaptive types will prefer to restrict their behavior to be socially acceptable (B agozzi & Foxall, 1995). While these subscores

PAGE 46

32 add further insight into how people solve pr oblems, they have been found to be less reliable with younger populations who are less mature (Kirton, 1999). Kirton (2003) has described ad apters as those who like to have structure in place when they are attempting to solve a problem Adaptors may appear cautious as they prefer to work within an established para digm of rules (Foxall & Bhate, 1993). They foresee problem-solving and decision making as a sound, thorough process (Foxall & Bhate, 1993). Adaptors are satisfied with devising a sm all number of sufficient solutions, and pursue efficiency in problem-solving by ma king steady progress towa rd a solution. Their cognitive behavior tends to be rule-governed in that they pr efer to conform as opposed to break rules (Foxall & Bhate, 1993). Adaptors more readily accept th e status quo and will not challenge the accepted, or try to change, way of tradit ionally doing things (Kwang, Ang, Ooi, Shin, Oei, & Leng, 2005). Kirton (1999) also discusses this, saying that they will work within the established theories, po licies and practices. Innovators, in contrast, are more likely to enjoy a l ooser structure as they go a bout solving problems (Kirton, 2003). The innovator will tend to offer more discontinuous solutions to problems while being seen as adventurous or as a risk taker (Foxall & Bh ate, 1993). Innovators have a tendency to strive for novelty, exploration, tria l and error, and risk-taking. Innovators will promote new understanding through profound procedural changes (Foxall & Bhate, 1993). Innovators are less likely to accept th e status quo and do not like to follow the accepted way of doing things (Kwang et al ., 2005). Kwang and colleagues (2005) found that adaptors and innovators will also prescrib e to different values when taking tasks into

PAGE 47

33 consideration. With respect to demographics, KAI scores for females tend to be more adaptive than males (Foxall & Haskins, 1986). Foxall and Haskins (1986) suggested that the KAI is a viable marketing tool for identification of consumer choices. In the ar ea of marketing, researchers have found that adaptors are attracted to products they know, are less tolerant of change, and are unwilling to explore (Foxall & Bhate, 1993). They prefer a reasoned argument in advertising as compared to persuasion (Foxa ll & Bhate, 1993). Adap tors tend to prefer the products they currently use and would solv e problems that arise from changes in that product. When searching for information, adaptors will seek and use information conservatively as they slowly work to a decision (Foxall & Bhate, 1993). Marketing and business researchers have found that innovators will use more sources of information to find solutions to a problem, and will trust discrepant advertising, as well as personalized advertising that encourages them to act impulsively (Foxall & Bhate, 1993). Innovators will seek information about more innovations th an adaptors, even when this information conflicts with current product use. Adaptors are content with products they currently use and will not necessarily go looking for new products (Foxall & Bhate, 1993). Research shows that Adaption-Innovation theory correlates with many personality traits that are related to consumers, but it also describes relationships between decisionmaking and problem-solving in consumers (F oxall & Bhate, 1993). Foxall (1996) found that the KAI is not a predictor of early a doption of new products. It was also concluded that innovators would require little persona l communication from marketers to adopt a new product. While adaptors want a lot of r eassurance from marketers, it does not matter

PAGE 48

34 if marketers went out of their way to pr ovide this information, as adaptors would eventually buy the new item either way (Foxall, 1996). Pershyn (1994), in a study of the KAI on natural creative processes, asked participants to recall a problem they faced a nd to draw the process in which they went about solving the problem successfully. It was found that high adaptors tended to draw a linear process in which they were orderly working through the process in fewer stages. High innovators, in contrast, showed a non-linear process in which they had random, complex approaches with more stages and in some cases no true end point. The KAI has been utilized in several studies to describe the psychology of computer usage. Foxall and Bhate (1991) f ound in a study of graduate business students that the number of computer applications utilized and dur ation of computer use were correlated with KAI scores. The researchers found that for home computer use, those who were highly innovative tended to use more than four package-based computer applications and show high personal involve ment with computing. Foxall and Bhate (1991) stated that there is a need for investigations of th e relationships with KAI and computer use. Foxall and Hack ett (1992) found in an investig ation with managers that use of software applications was positiv ely related to adaptive-innovative problemsolving style. They noted that the sufficiency of originality and rule conformity subscales were positively related to computer use, while efficiency was negatively related to computer use. This implies that in terms of computer use, sufficien cy of originality and rule conformity are relevant to innovative traits while efficiency is relevant to adaptive traits (Foxall & Hackett, 1992).

PAGE 49

35 Bhate (1999) used the KAI to examine c ognitive styles and different message sources on attitude change. Through a study of 15to 73-year-olds, it was concluded that it was too simplistic for advertisers to use one universal appeal (wheth er it is positive or negative) for all individuals. The decision -making process for adaptives was sourceoriented and was influenced by both positive and negative sources. However, innovators tended to rely more on negative sources as they felt positive sources were more time consuming. This implies that for informationdriven Websites designers need to consider more than just one type of design for a site. While the KAI was initially developed for use with adults with work experience, it has been found to be reliable with younger popul ations who are also affected similarly by style (Tefft, 1994; Kirton, 1999) Many university students have had experiences working that they can pull from when taking the i nventory (Kirton, 1999). Dependable KAI scores have been achieved with people as young as 15 y ears old; however, due to maturity levels it has been indicted that the sub scores of SO, E, and R should be ignored (Tefft, 1994). Taylor (1993) used the KAI for a group of 17to 18-year olds and noted that the theory worked with this younger audience in a sim ilar way as it does for adult populations. Taylor did discuss the need for explanati on of a few words in the KAI that were confusing to this youthful audience. Fisher, Macro sson, and Wong (1998) reported successfully using the KAI with undergraduat e students in engineer ing and business to test relationships between cogni tive style and team role pr eference. Foxall (1992) also reported successfully utilizing the KAI with a group of students enrolled in masters of business administration programs in the United Kingdom, Australia, and the United States.

PAGE 50

36 Information Richness The concept of information richness comes into play as people cognitively choose which media will help them as they work to solve problems. Individuals choose media they perceive to be most efficient in he lping them to complete a communication task (Kelleher, 2001). Based on equivocality or the ambiguity, the lack of clarity of information, users who are seeking informati on will choose either a rich or lean medium (Irani & Kelleher, 1997). Equivocality can be described as the existence of multiple interpretations and an organizationa l situation (Trevino, Daft, & Lengel, 1990). Trevino and colleagues further described equivocality by comparing an equivocal message to a Rorschach ink-blotmultiple users may read it differently than others depending on their unique backgrounds and pers pectives. Information tasks that are seen as unambiguous will be considered low in equivocality, while tasks that are based on processing of multiple interpretations may cause higher equivocality. The theory of information richness describes that users us e rational media choices to deal with these equivocalities (Irani & Kellehe r, 1997). Equviocality and ambi guity of a site can be tied back to many of the same underpinnings that drive adaptors and innova tors. This idea of something being equivocal or am biguous or rich or lean is similar to the way the KAI discusses structure. While more adaptive probl em-solvers will like structure, it could be argued that they too may like equivocal me ssages that are lean, and more innovative problem-solvers who work outside structure wi ll want more ambiguity in their messages that are rich. Media richness refers to a mediums tende ncy to present information in either a rich or lean manner. The richest comm unication medium is f ace-to-face, followed by the telephone and e-mail because these medi a allow immediate feedback and can be

PAGE 51

37 highly personal (Trevino, Daft, & Lengel, 1990). The leanest communication is formal documents such as fliers, bulletins, and qua ntitative reports. Iran i and Kelleher (1997) found in a study of equivocality that those in a high equivocality condition will choose a richer Website more often than a pamphlet or a lean Website. Based on this perspective, it could be assumed that as people work through the cognitive pr ocessing of their information-seeking tasks, not only their preferred st yle will come into play, but also the complexity of the task will influence the media choice. Uses and Gratifications Theory Uses and gratifications theory also addresses cognitive style as a motivator to fulfill a specific need (Stone, Singletary, & Rich mond 1999, p. 200) based on an individuals choice of media. The theory emerged in th e late 1950s and early 1960s as researchers looked to understand audience involvement in mass communication (Blumler, 1979). The Uses and Gratifications theory provided a re placement of the ideal that the audience member was a passive victim, and that one could actively look at media for their own purpose (Blumler, 1979). As one of the theories most associated with usage of media, Uses and Gratifications researchers define communication needs th at shape why people use media and the behaviors that gratif y those needs (Rubin, 1994). One such use that is noted in the literature is to find information or solve a problem (Katz, Blumle r, & Gurevitch, 1974). Uses and gratifications theorists assume th at communication needs interact with social and psychological factors to produce motiv ation to communicate (Rosengren, 1974). Katz, Blumler, and Gurevitch explain that the logical steps the theory is concerned with include (1) the social and psychological origins of (2) needs, which generate (3) expectations of (4) the mass media or other so urces, which lead to (5) differential patterns

PAGE 52

38 of media exposure (or engagement in other activ ities, resulting in (6) need gratifications and (7) other consequences, perhaps mos tly unintended ones (1974, p. 20). Five elements that have been described as assump tions to Uses and Gratifications research include 1) active audience, 2) audience memb er links need of gratification and media choice, 3) media compete with other sources for need satisfaction, 4) goals of mass media use can be derived from data provided by i ndividual audience members, and 5) value judgments about mass communication need to be suspended when audience orientations are explored (Katz, Blumler, & Gurevitch, 1974). Other assumptions associated with Uses and Gratifications approaches include a) media use is goal directed, b) media consumption can fill a wide range of need s, c) people have enough self-awareness to know and articulate their reasons for using the media, and d) gratifications have their origins in media content (McLeod & Becker, 1974). From a Uses and Gratifications standpoint the first assumption is that media users are active in their attempts to seek out a nd find information from the media channel of their choice (Rubin, 1994). The approach assu mes that users are active participants because they are active communicators who se lect which communication channel to use (Blumler, 1979). Theorists suggest that users will have some form of need, such as solving a problem, which they will try to gra tify with the use of a specific media (Baran & Davis, 2002). These motives and needs can be based on psychological characteristics, attitudes, and per ceptions (Rubin, 1994). As seen in the second assumption put forth by Katz, Blumler, and Gurevitch (1974), media fulfills one of four functions for individuals: it entertains them, serves as a mechanism for surveillance, correlates with what people know about society; or transmits

PAGE 53

39 society across generations (Baran & Davis, 2003). Based on these functions, media users will feel a need (needing to solve a problem, n eeding love or acceptance, or needing to be informed or entertained) that they will be motivated to gratify through a specific media outlet (Blumler, 1979). These needs lead someon e to actively seek out and use a specific medium that will in turn gr atify that perceived need (B ran & Davis, 2003). The medium individuals chose would be based on their expectations of that medium and their perceptions of how well it will gratify that need (Blumer, 1979) This theory assumes that communication behavior is sought out to fu lfill these cognitive needs by an individual user (Katz et al., 1974). The third assumption calls for researchers to realize that media compete for the ability to fulfill user needs (Katz, Blumler, & Gurevitch, 1974). Those needs served through the media and communication is only a small segment of the human needs that need fulfillment, and the degree to which t hose can be fulfilled through media varies (Katz, Blumler, & Gurev itch, 1974). Bouwman and Wijngaert (2002) called for researchers to take into account the persona l factors and situations that affect media choice, as was traditionally called for in this Uses and Gratifications approach. Uses and Gratifications and the Internet The Uses and Gratifications paradigm ha s been described as the best model in which to study new communication methods su ch as the Internet (Ruggiero, 2000). Many recent studies have utilized Uses and Gratifica tions as part of the theoretical framework when studying the Internet because of the interactivity, demassification, and asynchronicity it allows that other media do not (Baran & Davis, 2003). Such studies have shown that Internet use is motivat ed by the need to escape, the need for entertainment, the need for interaction, and th e need for learning and socialization (Baran

PAGE 54

40 & Davis, 2003). Papacharissi and Rubin (2000) examined audience us es of the Internet and identified five motives for using th e Internet including: information seeking, interpersonal utility, pass time, convenience, entertainment. Their findings also suggested that those who like to look ar ound the Internet felt it allowe d them to save money and obtain information (Papacharissi & Rubin, 2000). Through several case studies Bouwman and Wijngaert (2002) concluded that due to the fact that the receiver of information on the Internet has to seek it out, the assumption of an active audience in the Uses and Gratifications approach is greatly supported. Ko (2002) investigated whether motivations to use the Internet could explain key aspects of usage. Informati on, escape, time passage, and in teractivity were the four primary motivations to using the Internet discovered by the resear cher. Ko (2001) found that those who use the Internet for informa tion are more likely to satisfy their needs by using the Internet over other media. Web us ers are active information seekers, as they must click on links and use hypertext to navi gate online. Lin and Jeffers (1998) suggested that, in turn, Web use is goal directed as the users must be aware of the needs they attempt to satisfy. Ko, Cho, and Roberts (2005) looked at the Internet use of college students in the United States and Korea th rough an experimental design. Researchers concluded that consumers who possessed high information motives were more likely to engage in human-message interaction on a Website; while those with higher social interaction motives were more likely to engage in human-human interaction. The Uses and Gratifications approach ha s also been utilized to study political information posted online. Kaye and Johns on (2002) found that the four primary motivations for locating political informa tion online included guidance, information

PAGE 55

41 seeking/surveillance, entertainment, and so cial utility. It was found that guidance and information seeking/surveillance is linked to more purposeful uses of the Web than for just surfing. Previous Experience and Expectations As individuals choose the media that they will utilize to gratify their needs, they draw on memories of past media use to ai d them in that action (Katz, Blumler, & Gurevitch, 1974). The third step in the model, expectations, has been has been thoroughly researched through the Uses and Gratifica tions paradigm (Rayburn, 1996). Katz and colleagues (1974) allude to the expectations from media by the audience when selecting content to fulfill certain needs. The medium th at is used depends on a variety of factors, including the characteristics of the informa tion needed, the characte ristics of the person asking the question, and the context in whic h they have access to specific media (Bouwman & Van De Wijngaert, 2002). Peled and Katz (1974) found in a study of media during wartime and crisis that people came to a specific medium with expect ations of what that medium will be and what it will gratify for them. Bouwman and Wijngaert (2002) found in a study of characteristics of basic needs that certain th resholds of accessibility must be met before deciding to use a medium. One such threshold th ey describe is that of suitability, where the medium can provide them the inform ation they are searching for (Bouwman & Wijngaert, 2002). For the user to know if th at medium has that information, they must have some previous experience with it. Several theories have attempted to deal w ith theses expectations, such as Fishbein and Ajzens (1975) expectancy-value theory. This theory states that there are three kinds of beliefs: descriptive which is a result of di rect observation; information which is formed

PAGE 56

42 by accepting information from outside sources; and internal which include characteristics of the object that are not di rectly observed (Rayburn, 1996). This model has shown to be of great use to the study of Uses and Gratif ications where feedback loops are prevalent (Rayburn, 1996). Information Recall The Internet may not always be the best medium to reach audiences; Bogart (2000) reported that an experiment done at The Ohio State University showed that when readers were given an article in both print and We b versions, they report ed that the printed version was more understandable. A strength of the Web is its ability to present individual readers with a sele ction of tailored contents. Th is is also a weakness, if it means that they are no longer exposed to what they have not expected and did not know they wanted, (Bogart, 2000, p. 1). In contrast, DHaenens, Jankowski, and Heuvelman (2004) found in their study of two online and print versions of Netherlands newspapers that there was no evidence that online readers consume and retain news diffe rently than those r eading print versions. They found that online read ers recalled internatio nal news better than print readers. It was concluded that no evidence in the study suppor ted claims that online readers consume news differently from print readers. Moor e (2004) found in an experimental study of magazine and online advertisements that higher selective exposure was found for information online over the print version, a nd moderate recall differences were seen between the two media. Based on these findings Moore called for future research of new media to examine memory and media comparisons. Eveland and Dunwoody (2002) found that when compared to print media, the Web increases learning through an increased elaboration, but may decrease it through

PAGE 57

43 increased selective scanning. Tewksbury and Althaus (2000) found similar results with people reading online newspapers versus pr int additions. It was found that the online versions of the papers presented fewer clue s to the importance of events compared to print editions, and in turn people were more willing to use their own interest to guide what they focused on and were able to reca ll. Danaher and Mullarkey (2003) found that length of exposure to a site containing a banner advertisement affected how likely viewers were to be able to recall the inform ation. It was suggested by the researchers that designers should include inter active features that encourag e users to stay on a page longer. It was also noted that those in a goal-directed mode were less likely to remember information than those who were just surfing the Internet. While Till and Baack (2005) did not look at recall in terms of information online, they did add to the discussion by looking at the creativity of a dvertisements in terms of recall. It was discovered that in an unaided basis, creativ ity generates significantly more recall. Eveland and Dunwoody (2001) compared learni ng in print versus linear, nonlinear, and advisement Web designs. It was found that learning was better for print than nonlinear and linear; however, no diffe rence was found between print and an advertisement design (which included cues to work through the site such as back, next, and story map buttons). It was also noted that Web experts learned more than Web novices on all mediums. Wicks (1995) used an experiment looking at free recall, or re call not prompted, and extended recall, or recall after time has pa ssed, to see the effects of medium on news recall. Wicks found that indi viduals acquire common knowl edge from the news and that time is needed in the recall process.

PAGE 58

44 Several studies compared different forms of online media and other media to discover more about recall of informati on. Berger (2001) ascertained that those comfortable with hypertext did not have a sign ificant difference in re call than those with low comfort levels, concluding that presenti ng information linearly or nonlinearly would not offer users an advantage. Lowery (2002), however, found that linea rity has an effect on perceived control over the media experi ence, but did not lead to any increased knowledge. Fox and colleagues (2004) found in a study of television news that recall was greater for younger and older viewers when grap hics were present. Multimedia, such as video and imagery, has also been found to not increase comprehension or recall scores above those of a static, text -based site (Berry, 2001). Berry (2001) felt that multimedia may enhance a users recall of textual information if it was reinforcing the textual information on the page. Eveland and colleagues (2004) utilized an experimental design to discover how Web site organization influenced free reca ll of information. The researchers concluded that a nonlinear compared to a linear site ha d mixed results on learning. It was found that a linear site design increased factual learni ng in participants, but the nonlinear design increased the knowledge structure densit y. The learning of factual knowledge was hindered by the nonlin ear structure. Attitude Attitude has been shown to be an importa nt predictor of usage and implementation of technology (Rodgers & Chen, 2002). Eagly a nd Chaiken (1993) described attitudes as being derived to motivate behavior in orde r to exert effects at various stages of information processing. It is further c onceptually defined by the authors as a psychological tendency that is expressed by evaluating a particular entity with some

PAGE 59

45 degree of favor or disfavor (Eagly & Chaike n, 1993, p 1). An attitude is not formed until people are presented with a situation in which they must evaluate it on an effective, cognitive, or behavioral basi s (Eagly & Chaiken, 1993). While attitudes are not directly observable, they can be inferred from responses given that show some state or disposition that has been engaged (Eagly & Chaiken, 1993). Researchers have assumed that attitudes should be divided in to three classes cognitive, affective, and behavioral (E agly & Chaiken, 1993). The cognitive category contains all of the thoughts an individual has about the attitude object, while the affective category is the feelings and emotions one has in relation to the attitude object (Eagly & Chaiken, 1993). The behavioral category cont ains ones actions w ith respect to the attitude object. In cong ruence with the idea of three categories of atti tudinal responses, is the idea that there are three antecedents to attitude c ognitive processes, affective processes, and behavioral proc esses (Eagly & Chaiken, 1993). The cognitive process in which attitudes dr aw from is one in which much research derives (Eagly & Chaiken, 1993). The assumpti on by researchers is that attitude is derived from a cognitive learning process in which one gains information about the attitude object and then forms beliefs. The in formation is gained vi a direct and indirect experiences with the attitude object (Eagly & Chaiken, 1993). While research on attitudes has been mo re defined in the social psychology literature, it has been found in other social science literature as well (Eagly & Chaiken, 1993). Attitude research has been done extensively in the area of advertising and attitude toward ads and their effectiveness (Chen, 1999; Rodgers & Chen, 2002; Sinclair & Irani, 2005; MacKenzie & Lutz, 1989). Rodgers and Chen (2002) found adoption of the

PAGE 60

46 Internet by advertising agencies is affected by poor attitudes toward and lack of experience with Internet advertising. In 1999, Chen developed a scale, based on other evaluative scales, to provide re searchers the ability to measur e the attitudes of users about Websites to help indicate the value of such sites. Cho (1999) found in a study of advertisi ng on the Web that people who had more favorable attitudes toward the Web were mo re likely to click on advertising on a site. Rodgers and Chen (2002) looked at advertis ing in terms of the organization, and found that poor attitudes toward the Internet after adoption for advertising was due to the agencies lack of experience and expertise with that form of advertising. Teo and colleagues (2003) found that attitude toward a commercial Website can be positively influenced with increased interactivity on the site. Conceptual Framework Based on the literature presented in this ch apter, the conceptual framework seeks to explain a model in which: An individuals cognitive problem-solving style, when influenced by their level of previous usage of the media will affect their levels of attitude and information recall after being presente d with an interactiv e or non-interactive Website (Figure 2-1.).

PAGE 61

47 Figure 2-1. Conceptual fr amework for this study. Research Questions Based on the conceptual framework, this study will examine the relationship of cognitive problem-solving styles and level of Internet usage on perceived attitudes toward and information recall of sites that vary in their leve l of interactivity. This study will attempt to an swer the following questions: 1. To what extent does problem-solving style and Internet usage have in influencing perceptions of attitude and recall toward Websites that vary in level of interactivity? 2. To what extent is problem-solving style alon e a factor in influencing perceptions of attitude toward and information recall toward such a site? 3. To what extent is Internet usage alone a f actor in influencing pe rceptions of attitude and level of information recall toward such a site? Problem Solving Style ( More Ada p tor/Innovator ) Level of Usage Level of Interactivity of Site Recall High Attitude High Recall Low Attitude Low

PAGE 62

48 CHAPTER 3 METHODOLOGY Overview As Extension professionals continue to embrace online technologies to reach audiences and inform them about topics that will help solve problems with which they are faced, it is important that this information be presented in a form that will be usable, valuable, appropriate, and easy to recall. This method of disseminating Extension information is providing new challenges, as it is attempting to reach audiences in a new format that is usable and valuable. Thus, the purpose of this i nvestigation was to examine the effects of individuals problem-solving style, level of Website inte ractivity, and level of Internet usage on attitude and level of recall with respect to an information-driven Extension Website. Although usage has been looked at extensively in the literature it has never been tied to problem-solving style. By understanding how pr oblem-solving styles in particular affect users perceptions of a Website with respect to such attributes as attitude and information recall, Extension, agricultural communicator s, and commodity groups who are utilizing the Internet to reach audiences will be better able to utilize communications processes to inform audiences, educate them, and affect productive change. Hypotheses Based on the literature presented, th e following hypotheses were developed: H1: Unaided information recall and level of attitude toward an information-driven Extension Website will diffe r significantly as a function of site interactivity, problem-solving style, and level of Internet usage.

PAGE 63

49 For Subjects Who Receive the Interactive Site: H2: Attitude toward an information-driven Extension Website will differ significantly as a functio n of Internet usage and problem-solving style. H3: Unaided information recall will diffe r as a function of Internet usage and problem-solving style. For Subjects Who Receive the Non-Interactive Site: H4: Attitude toward an information-driven Extension Website will differ significantly as a functio n of Internet usage and problem-solving style. H5: Unaided information recall will diffe r as a function of Internet usage and problem-solving style. Research Design The research design for this study was expe rimental in nature. The design was a 2 (more adaptive/more innovative problem-solvi ng styles) x 2 (interac tive/non-interactive information-driven Extension Website) x 2 (hig h/low levels of Inte rnet usage) betweensubjects, factorial design focused on assess ing whether the problem-solving style of an individual, coupled with exposur e to an interactive or non-in teractive information-driven Extension Website and level of Internet usage, will influence information recall and attitude. Factorial designs allow for a more signi ficant test of hypothesis (Ary, Jacobs, & Razavieh, 2002) by determining the influence of an individual independent variable on another independent variable (Christenson, 2001). Factorial designs not only offer the ability to test more than one independent vari able, but they also allow for the testing of more than one hypothesis in one experiment (Christenson, 2001). Beyond testing just the independent influence of a specific variab le, factorial design a llows for tests of interaction to be performed (Christenson, 2001). In order to generate a large enough sample to effectively test the hypotheses, a minimum sample size of 180 (2 x 2 x 2 x 30=

PAGE 64

50 240) is needed to ensure at least 30 subj ects for each condition (Christensen, 2001). The group design appears as follows (Gall, Borg, & Gall, 1996): R X1 O1 R X2 O1 In this posttest-only, randomized su bject design: R= random assignment, O1= posttest measures, X1= interactive version of the site, X2= non-interactive version of the site. The basic threats to internal validity were considered in the design of the study. As identified by Campbell and Stanley (1963), th reats to be cognizant of include history, maturation, testing, instrumentation, regr ession, subject select ion, mortality, and interaction effects. The posttest only design of this st udy was expected to address regression, history, and maturation (Campbell & Stanley, 1963). Mortality was a concern in this study as students were asked to give data on two di fferent occasions, once in the classroom and once online. Time between thes e two administrations was less than a day. To address this concern, data was examined post collection, and showed that this was not an issue since only 10 participants did not complete both sections. Interaction of extraneous variables was also a possibility. To address this concern factorial design was utilized. A factorial design allows for the confounding variables to be built into the design. Confounding variables that were cont rolled for included gender (Kirton, 2003), time, content (Saunder, 2000), and previous experiance. To address instrumentation validity, a panel of experts was utilized to l ook at the face validity of the items, and a pilot test was run to ensure construct va lidity. The pilot study included testing of the instrument as well as the messa ge stimulus. Threats due to testing are described as the

PAGE 65

51 effects of taking a first test upon the scores of a second te st (Campbell & Stanley, 1963). This could be a concern as the participants took two different inst ruments at different times. However, the instruments were on differe nt topics and were not in a pre-test/posttest situation where they w ould affect one another. While the undergraduate cour ses, in which the sample was derived, were selected through availability, to counteract any valid ity threats posed by the selection of participants, the version of the sites were ra ndomly assigned to the participants and data was collected on each individual student and not the course as a whole. Manipulation checks were conducted on the version of the sites to ensure no differences were found. These manipulation checks will be described in more detail later in this chapter. Threats to external validity were also take n into consideration dur ing the design of the experiment. Threats which need to be a ddressed, as outlined by Ary and colleagues (2002), include population validity, ecological validity, and experimenter validity operation. The threat of population validity must be taken into account as the population consisted of students in three undergraduate courses at a large Southeastern land-grant university. While this sample is not generali zable to the whole population, the majority of Internet users are in the 18to 30-year age range (CyberStats, 2005) and are a viable population to study in terms of Internet modality Students familiarity with the Internet aids in assuring differences between exposure groups is unlikely due to the novelty inexperience of the Internet (Tewksbury & Althaus, 2000). Bull a nd colleagues in 2004, called for Extension to pay specific atte ntion to underserved audiences who have typically not been an original stakeholder in the program.

PAGE 66

52 Ecological validity describes how generalizab le the experimental environment is to other environments by taking into account pretest and post-test sensations, multi treatments, Hawthorne effects, novelty effect s, and experimenter effects (Ary et al., 2002). To address these threats, students were given science-genera ted content that is viable and interesting to them. Questions were asked to ensure this topic was of interest during the study. To help ensure a more natu ral environment when taking the instrument, students were asked to take the second portion of the study at home on their own computers. By doing this, subjects were in a familiar setting and will be looking at a topic in which they might research on their own tim e. This also helped to curb any novelty effects that could occur during the treatment. Subjects Participants were recruited out of two service courses taught in the College of Agricultural and Life Sciences at a large Southeastern land -grant University. The courses serve as part of a general education require ment for students across the university and are thus taken to be largely representative stude nt population with a variety of majors and backgrounds. A total of 314 students comple ted the initial usage and problem-solving instruments through direct administration. T hose students who were enrolled in more than one of the courses utilized in the study were instructed to pa rticipate in the study only once. Their names were noted during the firs t data collection to insure that they did only participate once. In order to manipulate the treatment ve rsion of the site, subjects were randomly assigned to either an interact ive or non-interactive version of the same Extension Website. Manipulation checks were completed during the pilot study to insure that there were no significant differences among the two courses or those receiving one version of the site over the other (Table 3-1).

PAGE 67

53 Table 3-1. Independent Sample T-test for Significant Diffe rences Between Courses and Version of the Site Based on Age or Gender. T df Sig. Mean Diff. Courses Gender .80 253 .43 .26 Age -.79 254 .43 -.06 Version of the site Gender 1.27 253 .20 .41 Age 1.1 254 .28 .08 Pilot Study To establish reliability and validity of the final instrument, a pilot study was conducted with 29 undergraduate students in an agricultural leadership course at a large Southeastern university. Care was taken to ensure no students who participated in the pilot were a part of the final sample. The same procedures were utili zed in the pilot study as they were for the full study, as outlined below. Prior to the pilot test, a panel of experts assessed the face and content valid ity of the instrument. To a ssess construct validity, item analysis was run on the pilot instrument. An overall Cronbachs alpha reliability of .72 was computed. A few adjustments were made to refine and finalize the instrument. Message testing was also completed during the pilot study. Particip ants in the pilot section were cross-referenced w ith participants in the full study via the reported emails to ensure no subject participated in both data collections. Manipulation checks were conducted duri ng the pilot study to assess whether respondents could distinguish between the trea tment and control version of the Websites. Subjects in the treatment and control version of the sites were asked to identify if the version of the Website they we re presented with was inter active. An independent-sample t-test was conducted using versi on of the site as the indepe ndent variable. Results show that participants receiving the interactive version of the Website were successfully able to

PAGE 68

54 identify that the version of the site they vi ewed was interactive. Those receiving the noninteractive version of the site were also able to identify that their version of the site was non-interactive. A significant difference was found in the response between the condition groups. Their response is presente d in Table 3-2 and Table 3-3. Table 3-2. Means Table for Website In teractivity Identification (N=19*). Page is Interactive N Mean SD Interactive Version 10 3.00 1.50 Non-interactive Version 9 1.44 .73 *19 out of 29 pilot study part icipants completed the manipulation check. Using a 1-5 Likert Scale (1= Strongly Di sagree and 5= Strongly Agree) Table 3-3. Independent Sample T-test for Significant Differences Between Less and Interactive Version of the sites. T df Sig. Mean Diff. Non-interactive/interactive Version -2.84 17 .011 -1.56 Procedure The instruments for both the pilot and full studies were administered in two parts due to the length of instruments and the use of a standardized instrument (KAI), which could only be administered on paper. The first part of the instrumentation was administered in the classroom. During this time a certified KAI representative administered the KAI to the students.1 Participants were also given questions on basic Web and media use, attitudinal scales on per ceptions of the Internet, and demographics. Upon completing the first part of the instrumentation, participants were asked to report a university email. Students were informed that they would be contacted later that day via email with the second part of the instrumenta tion. They were informed that the course 1 Dr. M.J. Kirton, director and founder of the Occupational Research Center, developed this psychometric inventory. KAI administrators must complete an intensive weeklong course given by Dr. Kirton on the KAI instrument and the underlying theory in order to be certified to administer the instrument.

PAGE 69

55 title would serve as the subject to the email, to ensure students did not pass the email off as spam. Once students were randomly assigned to ei ther the treatment or control, part two containing a link to the appr opriate version of the site was emailed (Appendix B). Participants could only view one of the tw o versions based on the random assignment. Once at the appropriate version of the site, participants view ed a consent screen and were directed to click on a link to open a new browser window containing the information page of the site (Figure 3-1) Participants were instructed to spend as much time as needed to review the content before completing the final instrument. Figure 3-1. A screen capture of the consen t information and instructions sent to participants. After exposure to the version of the site, pa rticipants were inst ructed to close the browser window containing the information page and not to return to it. They were then directed to click on a link which took them to a new page containing part two of the instrument in which they were asked their attitudes toward the treatment or control version of the site to which they were e xposed, were asked to recall information and report knowledge on the topic presented, and we re asked about their current and past

PAGE 70

56 usage of Extension information through an online form. After data collection was completed, the data from the part one instru ment and part two inst rument were matched based on an email identifier, which was included in both sets of instruments. The final subjects were solic ited out of two large servic e courses taught at a large Southeastern university (N=314). A total of 314 students completed part one of the instrumentation in the class. A total of N=305 individuals completed both parts of the study. One of the instruments was then thro wn out for being incomplete. Another 48 instruments were thrown out due to the se nsitivity of the KAI instrument, leaving 256 total participants for the full study.2 Instrumentation Instruments for the first part of th e study consisted of the Kirton AdaptionInnovation Inventory (32 items), a 49-item inst rument measuring media usage, (7 items) Internet experience (17 items), attitude scales toward the Internet (11 items), a scale on the value of the Internet (8 items), and demographics (6 items). (See Appendix A.) The instrument in the second part consisted of 34 items measuring attitudes toward the interactive or non-interactive ve rsion of the site to which th ey were exposed (11 items), information recall (4 items), Extension usage (2 items), and knowledge and interest in the car-buying information (10 items). (See Appendix B.) 2 The KAI is a very sensitive instrument in which partic ipants who respond that everything is easy or hard for them or those who select down the middle of the scal e must be rejected as it is suspected that they are being reluctant to respond truthfully or are trying to deliberately sc ore in an acceptable way (Kirton, 1999, pp.19). It is noted by KAI researchers (Tefft, 1994; Kirton, 1999) that younger populations will have lower maturity levels, affecting the rejection rate of such a population. However many university students have enough work experience to understand the items without problems (Kirton, 1999). A 10%-20% wastage rate can be expected un der favorable conditions with university students (Kirton, 1999).

PAGE 71

57 Independent Variables Treatment The only independent variable that was mani pulated in this study was the level of Webpage interactivity to which the subj ects were exposed. The other independent variables were measured on the basis of data collected via instrumentation. For the purposes of this study, two vers ions of a Webpage were created: an interactive version and a non-interactive ve rsion (Appendix C.). The versions were created using the same information; only the ability to interact with the message was manipulated (Sicilia, Ruiz, & Munuera, 2005) The non-interactive version was set up like a traditional Extension fact sheet, that ar e typically made available in PDF or basic HTML formats online, where the entire message was on one Webpage with no interactive elements (Figure 3-2). The Extension logo was visible, and bolded headlines broke up the text. Figure 3-2. A screen capture of the non-intera ctive control page sent to participants.

PAGE 72

58 The interactive version contained the same information and Extension logo presented in a Macromedia Flash format (Figure 3-3). After the Extension logo an animated car could be seen moving across the screen. An introduction and instructions to click along the car-buying path followed. User s could then click on the icons along the car-buying path and a pop-up window would app ear with information that could be moved around the screen or closed. Figure 3-3. A screen capture of the interactive version of the site page sent to participants. The information content of the page was consistent across both stimuli conditions. The information used was adapted from an Extension fact sheet produced by the University of Iowa (1998). To make it relevant and salient to the subj ects, the topic of car purchasing was chosen as the focus of the in formation presented on the page used for the

PAGE 73

59 treatment and control version of the sites. This was due to ex pectation that the information would be salient to this audien ce. College students are in a unique situation in that they are becoming more important to automakers as a demographic which will soon be making car purchases (Clemens, 2005; Collier, 2006). These echo boomers are at the prime age to buy their first car and ar e said to spend more of their income on products, such as automobiles, than others (Clemens, 2005). Market research has shown that these young, first-time buye rs are turning to the Intern et as a main source when making car-buying decisions (Associated Press, 2006). As individua ls begin looking at buying large items like a car they will move into a problem-s olving mentality of trying to decide what type of car is best for their needs. Problem-solving style As described in Chapter 2, all people are bound by their makeup as to how they define and solve the problems with which they are faced (Kirton, 1999). In this study, the Kirton Adaption-Innovation Inventory (KAI) was utilized to place individuals on a continuous scale between being adaptive and innovative in th eir problem-solving style. This cognitive style is a tra it that can be expected to be stable over time and across situations (Kirton, 2003). It was assumed problem-sol ving style would influence preference for and recall of information needed to solve a problem, such as how to purchase a car. The KAI inventory requires respondents to as sess the degree of ease or difficulty they encounter in sustaining their adaptive and innovative behaviors over periods of time by drawing an x where they f it in a series of 32 five-poi nt scaled items ranging from very easy to very hard (Foxall & Bhate, 1993) (See Table 3-4). Individual scores are composed of three independent sub-scales which measure originality (13 items),

PAGE 74

60 efficiency (seven items), and rule-confo rmity (12 items) (Goldsmith, 1984). Responses are computed into overall scores rang ing from 32 to 160 (Foxall & Haskins, 1986). Respondents scoring below the 96 mid-point are considered adaptors, while those above 96 are innovators (Foxall & Haskins, 1986). The KAI inventory has been shown to have a high level of internal consistenc y; Kirton (1999) returned a Kuder-Richardson 20 reliability of .88 and then retested with a similar popula tion a year later to again receive a K-R 20 of .88. In The KAI Manual it is reported that 31 studies from 12 countries have yielded Cronbachs alpha s ranging from .79 to .91 (Kirton, 1999). Goldsmith (1984) reported a reliability Cronb achs alpha of .84 for the KAI while Goldsmith and Matherly (1986) reported a Cronbachs alpha of .87. Both studies were run with undergraduate populations. Table 3-4. Example of KAI Instrument Items How easy or difficult do you find it to pres ent yourself, consistently, over a long period of time as: Very Hard Hard Easy Very Easy A person who is patient A person who conforms A person who enjoys the detailed work. Internet Usage As discussed in Chapter 2, communicati on needs interact with social and psychological factors to produce motivati on to communicate (Rosengren, 1974). These factors and previous experiences with a medi a will influence a user to choose a specific media to gain information. Ko (2001) f ound that those who us e the Internet for information are more likely to satisfy their needs by using the Internet over other media. Web users are active information seekers, as they must click on links and use hypertext to navigate online. College students represen t the largest population of Internet users

PAGE 75

61 (Eastin, 2001). Large percentages, 96%, of all 18to 29-year-old user s find the Internet a good way to get information (Fallows, 2004). Level of Internet usage is thus important to understand and gauge. For the purposes of this study, Internet usage was defined by the amount of Internet use each week, the number of sites subjects visit, and the activities they perform while online. In this study le vel of usage was meas ured through a 13-item researcher-developed scale (Appendix A). In order to measure subjects usage of and experience wi th the Internet, subjects were asked several researcher-developed questions about how many hours they spend online each day and how many sites they vi sit in an average session. Respondents were asked to rank on a five-point Likert scale how often they participat e in 10 specific online activities such as downloading musi c and shopping online (Table 3-5). Table 3-5. Example of Internet Usage Items. Please indicate how often you do the following each week Download music 1 Never 2 3 Sometimes 4 5 Very Often Read a blog 1 Never 2 3 Sometimes 4 5 Very Often Instant message 1 Never 2 3 Sometimes 4 5 Very Often Read Facebook or MySpace 1 Never 2 3 Sometimes 4 5 Very Often Watch videos 1 Never 2 3 Sometimes 4 5 Very Often Download RSS (Real Simple Syndication) 1 Never 2 3 Sometimes 4 5 Very Often Shop online 1 Never 2 3 Sometimes 4 5 Very Often Shop/sell on EBay 1 Never 2 3 Sometimes 4 5 Very Often Use a search engine 1 Never 2 3 Sometimes 4 5 Very Often Work on WebCT or other online course 1 Never 2 3 Sometimes 4 5 Very Often How often do you use the Internet to find news/information online 1 Never 2 3 Sometimes 4 5 Very Often How many hours a week do you spend on the Internet 1 or less 2-3 4-5 6-7 8-or more How many sites do you visit on an average session online 1-2 3-4 5-6 7-8 9 or more

PAGE 76

62 Dependent Variables Information Recall Several researchers have looked at how di fferent components and interfaces online affect information recall; however, after a thorough literature review, no studies were found that have examined how this is aff ected by trait variables such as cognitive problem-solving. Danaher and Mullarkey (2003) have found that length of exposure to a Website containing a banner advertisement a ffected how likely viewers were able to recall the information. Berger (2001) discove red that those subjects comfortable with hypertext did not differ significan tly in recall than those with low comfort levels. Lowery (2002) found that linearity has an effect on perceived control over the media experience, but did not lead to a ny increased knowledge. Information recall has been typically measur ed by asking participants to engage in free or unaided recall followed by a set of aided recall questions (DHaenens, Jankowski, & Heuvelman, 2004; Davis et al., 2005; Danaher & Mullarkey, 2003). Strong correlations have been reported by research ers utilizing both unaided and aided recall (Davis et al., 2005). Other re searchers comparing the differe nces between free or unaided and cued or aided recall found no significant di fferences in the findings (Padilla-Walker & Poole, 2002). However, while they are related, it has been noted in psychology literature that they represent different tasks (Padilla-Walker & Poole, 2002). It has been discovered that aided recall can cause more r ecollection of weaker memories, thus giving less accurate results (Padilla-Walker & Poole, 2002). For the purpose of this study, the learning process and free recall of informati on is of interest, ra ther than recognition. Thus, participants were asked, after reviewi ng the interactive or non-interactiv e version

PAGE 77

63 of the message structure to which they we re exposed, to recall information through an unaided response where they listed all information recalled from the site (Davis et al., 2005; Eveland, Cortese, Park, & Dunwoody, 2004; DHaenens, Jankowski, & Heuvelman, 2004). All true statements were scor ed as a +1 while untrue statements were coded as a -1 (Davis, et al, 2005). Points were summed to attain a mean unaided recall for the information contained in the version of the site to which they were exposed. Attitude Attitude has been shown to be an importa nt predictor of usage and implementation of technology (Rodgers & Chen, 2002). While a ttitudes are not directly observable they can be inferred from responses given that show some state or dispos ition that has been engaged in (Eagly & Chaiken, 1993). The assump tion by researchers is that attitudes are formed through a cognitive learning process wh ere one gains information and then forms beliefs. The information is gained through e xperiences with the object, such as the Internet or a particular We bsite (Eagly, & Chaiken, 1993). The most common way to measure attitude is through semantic differentials (Eagly & Chaiken, 1993). During the development of th is measure, researchers have found that three factors are usually underlying the scales : evaluation, potency, a nd activity (Eagly & Chaiken, 1993). The evaluative factor accoun ted for the most variability among scale ratings analyzed and was identified to repres ent attitude. The bipolar-adjectives that load in the evaluative dimension, like good/bad a nd pleasant/unpleasant, are thus used in semantic differentials to measure attit udes (Eagly & Chaiken, 1993). Two researcherdeveloped semantic differential scales were thus utilized. Att itude toward the treatment or control version of the site to which subjects were exposed (Tab le 3-6) and the Internet in general (Table 3-7) was test ed through two sets of 11 se mantic differential scales

PAGE 78

64 (good/bad, pleasant/unpleasant, trustworthy/untrustworthy, effective/ineffective, useful/not useful, and favorable/unfavorable ) (Sicilia, Ruiz, & Munuera, 2005). These bipolar adjectives were placed at each end of a five-point scale. Three out of the eleven attributes were reverse coded to decreas e the influence of response layout (Dillman, 2000). Table 3-6. Example of Scale Used to Measur e Attitude toward the Treatment or Control Version of the site to Which Subjects were Exposed The information presented on this Website is Good 1 2 3 4 5 Bad Not credible 1 2 3 4 5 Credible Biased 1 2 3 4 5 Unbiased Difficult to understand 1 2 3 4 5 Easy to understand Important 1 2 3 4 5 Not important Not interactive 1 2 3 4 5 Interactive Easy to find 1 2 3 4 5 Hard to find Not beneficial 1 2 3 4 5 Beneficial Believable 1 2 3 4 5 Unbelievable Not trustworthy 1 2 3 4 5 Trustworthy Accurate 1 2 3 4 5 Inaccurate Measure of attitude toward the Internet in general was measured for descriptive purposes only. Table 3-7. Example of Scale Used to Measure Attitude toward the Internet in General I feel that many Website s on the Internet are Good 1 2 3 4 5 Bad Credible 1 2 3 4 5 Not credible Unbiased 1 2 3 4 5 Biased Difficult to understand 1 2 3 4 5 Easy to understand Not important 1 2 3 4 5 Important Not interactive 1 2 3 4 5 Interactive Easy to find 1 2 3 4 5 Hard to find Beneficial 1 2 3 4 5 Not beneficial Believable 1 2 3 4 5 Unbelievable Trustworthy 1 2 3 4 5 Not trustworthy Accurate 1 2 3 4 5 Inaccurate

PAGE 79

65 An eight-item scale measuring the importance of the Internet in subjects lives was also utilized for descriptive purposes (Table 3-8). This index adapted by Ko (2001) from Rubin (1985) and Conway and Rubin (1991) asked subjects to indicate level of agreement with statements that discuss the im portance of Internet in their lives. The index has a reported internal reliability Cronbachs alpha of .86 (Ko, 2001). Table 3-8. Example of Index Used to M easure the Importance of the Internet Please rank your level of agreemen t with the following statements I would rather surf the Internet than do something else. 1 Strongly Disagree 2 3 4 5 Strongly Agree My knowledge increases as my Internet usage increases 1 Strongly Disagree 2 3 4 5 Strongly Agree It would be very difficult for me to survive without the Internet for several days. 1 Strongly Disagree 2 3 4 5 Strongly Agree Internet users are better educated people. 1 Strongly Disagree 2 3 4 5 Strongly Agree The Internet opens doors that would otherwise be closed. 1 Strongly Disagree 2 3 4 5 Strongly Agree Information online should be engaging. 1 Strongly Disagree 2 3 4 5 Strongly Agree Information online should be interactive. 1 Strongly Disagree 2 3 4 5 Strongly Agree Information online should be entertaining. 1 Strongly Disagree 2 3 4 5 Strongly Agree Data Analysis The data analysis for this study was completed using SPSS 12.0 for Windows PC. Multiple analysis of variance (MANOVA) was util ized to allow for a more sophisticated analysis of multiple independent and depe ndent variables (Graziano & Raulin, 2000). MANOVAs allow for more complex examinati ons of the simultaneous relationships of many variables, allowing researchers to crea te more sophisticated models to explain

PAGE 80

66 social behaviors (Sweet & Grace-Martin, 2003). Several mu ltivariate analyses of variances were then used to compare means and interaction effect s. Effect sizes of univariate analyses of variances were calcula ted to describe the ma gnitude of treatment effect (Kotrlik & Williams, 2003). Effect size reporting allows for judgment on the magnitude of differences between groups, and allows for better comparison to previous research results (Kotrlik & Williams, 2003).The Chohens f, which estimates the proportion of variance explained for the sample by the categorical variable was calculated as follows: (Kotrlik& Williams, 2003) 2 = SSbetween/SSTotal Cohens f = Square root of ( 2/12)

PAGE 81

67 CHAPTER 4 RESULTS The purpose of this study was to examine th e effects of problem-solving style, level of Website interactivity, and Internet usage on an individu als attitude toward an information-driven Extension Website and su bjects recall of the information presented on that site. Based on a conceptual fram ework relating Kirtons Adaption-Innovation (KAI) theory and the theory of Uses and Gr atifications, research hypotheses were formed with attitude and information recall as the dependent variables. The instruments and experimental cond ition were administered to a sample (N=314) of college undergraduates at a larg e Southeastern university. A total of 314 instruments were distributed in class. Those 314 participants were th en sent either the treatment or control condition and final instrument. A total of 305 participants returned the final instrument for a 96.8% response. Cases were then removed based on the following criteria: 1. The respondent had not fully comp leted all the instruments (n=1) 2. The respondent indicated on the KAI that nothing was easy or hard for them, indicating their score was suspect (n=48).1 1 The KAI is a very sensitive instrument in which participants who respond that everything is easy or hard for them or those who select down the middle of the scale must be rejected as it is suspected that they ar e being reluctant to re spond truthfully or are trying to deliberately score in an accepta ble way (Kirton, 1999, p. 19). It is noted by KAI researchers (Tefft, 1994; Kirton, 1999) that younger po pulations will have lower maturity levels, affecting the rejection rate of such a population. However, many university students have enough work expe rience to understand the items without problems (Kirton, 1999). A 10%-20% wastage rate can be expected under favorable conditions with university students (Kirton, 1999).

PAGE 82

68 This resulted in a fina l N of 256 participants. Demographics General demographics were calculated fr om the sample for gender, age, and college rank (Table 4-1). There were 110 males (43.1%) and 145 female (56.9%) respondents. The majority of respondents were 18-20 years old (56.3%), followed by respondents 21-23 years old (37.9%), re spondents 24-27 years old (5.1%), and respondents 28 years or older (0.8%). Th ere were 119 (46.9%) who reported being college juniors, 66 (26%) sophomores, 61 (24%) seniors, and 8 (3.1%) freshmen. Table 4-1. Number of Respondents by Age, Gender, and Class Rank Characteristic N % Age (n=256) 18-20 144 56.3 21-23 97 37.9 24-27 13 5.1 28+ 2 .8 Gender (n=255) Male 110 43.1 Female 145 56.9 Rank (n=254) Freshman 8 3.1 Sophomore 66 26.0 Junior 119 46.9 Senior 61 24.0 The majority (73.1%, n=187) of respondent s indicated being enrolled in the College of Agricultural and Life Sciences, followed by 9.8% (n=25) in the College of Health and Human Performance, 4.7% (n=12) in the College of Business, 3.7% (n=10) in the College of Liberal Arts and Sciences, 3.5% (n=9) in the College of Public Health and Health Professions, 1.6% (n=4) in the College of Design and Construction Planning, .8%

PAGE 83

69 (n=2) in the College of Pharmacy, and .4 % (n =1) in the College of Engineering and the College of Medicine, respectively (Table 4-2). Table 4-2. Number of Respondents by College (n=252) College N % College of Agricultural and Life Sciences 187 73.1 College of Health and Human Performance 25 9.8 College of Business 12 4.7 College of Liberal Arts and Sciences 10 3.7 College of Public Health and Health Professions 9 3.5 College of Design and C onstruction Planning 4 1.6 College of Pharmacy 2 .8 College of Engineering 1 .4 College of Medicine 1 .4 Undecided 1 .4 Media Selection and Internet Usage When asked to indicate their prefe rred choices of media when seeking information/news, 60.9% (n=156) indicated th ey preferred to use the Internet, while 26.6% (n=68) preferred television, 8.2% (n =21) preferred newspapers, 2.0% (n=5) preferred magazines, 1.2% (n=3) preferred radio, and 1.2% (n=3) preferred information from a book. Participants were asked to de scribe their Internet and co mputer usage. The majority (98.8%, n=253) indicated that they own a pe rsonal computer. High speed (55.3%, n=140) and wireless access (37.5%, n=95) were the mo st indicated methods to access the Internet at home, while at school the majority use a computer lab (49.8%, n=126) (Table 4-3).

PAGE 84

70 Table 4.3. How Participants Access the Internet at Home and at Campus Access n % At Home (n=252) High-speed 140 55.3 Wireless 95 37.1 Dial-up 16 6.3 Computer lab 1 .4 At Campus (n=253) Computer lab 126 49.8 High-speed 65 25.7 Wireless 62 24.5 Respondents identified whether they personally had a Web log (blog), Facebook page, MySpace page or a Website. The major ity (89.5%, n=229) did not have their own Website, and 89.1% (n=228) did not have a personal blog, while 85.2% (n=218) had a page on Facebook, and 10.5% (n=27) had a page on MySpace. Of respondents, 82.4% (n=210) indicated they had never created a Website. Participants indicated their level of attitude toward th e Internet in general through semantic differentials. The Internet was s een to be moderately good, easy to understand, important, easy to find, beneficial, believable and accurate (Table 4-4.). The grand mean for general attitude toward the Internet was 3.2 (SD= .91) on a 1 to 5 scale (1 being negative and 5 being positive). Table 4-4. Level of General Attitude Toward the Internet n M SD Beneficial 256 3.7 1.0 Easy to Understand 255 3.6 1.0 Easy to Find 255 3.6 1.1 Good 256 3.5 .90 Interactive 254 3.3 .97 Important 254 3.1 1.0 Believable 256 3.1 .75 Accurate 256 3.1 .77 Credible 256 3.0 .80 Trustworthy 255 2.9 .76 Unbiased 255 2.4 .90

PAGE 85

71 Extension Usage Questions were asked on a 1 to 5 likert scal e (strongly disagree to strongly agree) to determine if subjects had experiences w ith Extension information. A mean of 1.96 (n=254) indicated that the major ity of participants have not used Extension information in the past, demonstrating that th ese participants were not heav y users of Extension services in general. A mean of 2.11 (n =253) indicated that these part icipants have not visited the University of Floridas Extension We bsite in the past. (See Table 4-5.) Table 4-5. Mean Extension Expe rience of Study Participants* n M SD I have used Extension Information 254 1.96 1.21 I have looked at an Extension Website 253 2.11 1.40 *Based on a 1-5 Scale (1= strongly disagree to 5= strongly agree) Message Relevance When asked how important the informa tion presented to them was, 160 (62.5%) indicated that this information was moderate ly important to very important to them. A total of 150 (58.6%) of the responde nts indicated that they were moderately interested to very interested in the in formation on car buying. The majority (50%, n=128) were moderately knowledgeable on th e topic of car buying. The ma jority (74.2%, n=190) have not recently purchased a car, but 56.0% (n=141) have thought of purchasing recently. In general, the majority of part icipants were interested in and moderately informed about purchasing a car in the near future. Manipulation Checks Manipulation checks were conducted to evalua te the independent variables used in the study. Based on the literature and findings from the pilot study, two versions of a Website were developed contai ning the same information. Both versions contained facts on car buying and an Extension identificati on image, but differed in the amount of

PAGE 86

72 interactivity offered to the respondent. While car buying is no t a traditional agriculture message, the information presented was developed by agricultural economists and presented through Extension. To assess the face a nd construct validity issues, participants exposed to both versions were asked to iden tify along a five-point Likert scale if they strongly agreed to strongly disa greed that the site version th ey viewed was interactive or not. The means for both groups indicated that overall, on a Likert scale ranging between 1 and 5, they correctly identified the version th ey viewed as being either interactive or non-interactive (Table 4-6). An independent sample t-test was run to test for the significance of the condition manipulation with the version of the si te serving as the independent variable. Results showed a signi ficant difference between the treatment and control version of the sites at a .05 alpha level (Table 4-7). Table 4-6. Means Table for Site Interactivity Identification* Site is Interactive N Mean SD Interactive Site 123 3.35 1.22 Non-interactive Site 132 1.96 1.11 *Based on a 1-5 scale (1= strongly disagree to 5= strongly agree) Table 4-7. Univariate Analysis of Varian ce for Significant Differences between NonInteractive and Interactiv e Version of the Sites. df MSE F p Non-interactive/interactive 1 1.37 86.55 .000 Problem-solving Inventory The problem-solving instrument (KAI) used in this study, as described in Chapter 3, included 32 items. Respondents were asked to indicate the degree of ease or difficulty they encountered in sustaini ng their adaptive and innovative behaviors over periods of

PAGE 87

73 time by drawing an x where they fit in a five -point scale ranging from very easy to very hard. Scores ranged from 50 to 133 with a mean score of 92.6. As with many other variables in psychol ogy (Graziano & Raulin, 2000), the KAI is reported as a continuous score, which are often preferred by st atisticians for the ability to run simpler calculations (Agresti & Finlay, 1997). Grouped distributions are ge nerally required when work ing with a continuous variable such as KAI when reporting dem ographics (Graziano & Raulin, 2000). When there are many possible scores the data reporte d in tabular form will be long and difficult to read (Graziano & Raulin, 2000). Thus, a m eans split was conducted on the continuous variable for descriptive re porting purposes only. The Theory of Adaption-Innovation is deeply based on the idea that cognitive pr oblem-solving style measured by the KAI is based on a continuous scale, and should be treated that way in complex statistical interpretations (Kirton, 1999). In order to fo rmulate the means split, participants with scores over 96 were considered to be more innovative, and those below 96 were deemed more adaptive (Kirton, 2003). The means split resulted in 115 innovative participants (M=105.34, SD= 7.64) and 141 adaptive particip ants (M=82.18, SD= 12.51) (Table 4-8). Table 4-8. Means Table for Problem -solving Style Based on the KAI. N (N%) Mean SD Adaptive 141 (55.1%) 82.18 12.51 Innovative 115 (44.9%) 105.34 7.64 Descriptive statistics were run on the categ orizations of being more adaptive or innovative based on the literature that fema les tend to be more adaptive than males (Foxall & Haskins, 1986). This study supported the literature with more males (N=58) who were more innovative than females (N=5 1) and more females (N=93) than males

PAGE 88

74 (N=52) who were more adaptive in their prob lem-solving style (Table 4-9). The overall mean KAI score for female participants in this study was 89.42 (SD=16.84) and for male participants, 96.57 (SD=13.07). Table 4-9. Means Table for the Effect of Gender on Problem-solving Style Based on the KAI. N Mean SD Adaptive Male 52 85.94 8.89 Female 93 80.67 13.86 Innovative Male 58 106.10 7.78 Female 51 105.37 7.45 Internet Usage Constructs As indicated in Chapter 3, a 13-item researcher-developed construct was developed to assess Internet usage. Respondent s were asked to indica te on a five-point Likert scale how many hours th ey spent online each day, how many sites they visited in each stint online, if they ha ve ever created a Website, and how often each week they downloaded music, read a blog, instantmessaged, read Facebook/MySpace, watched online videos, shopped online, used search en gines or WebCT, and accessed news online. Standard deviations for the scale ranged between .9 and 1.5, indicating a satisfactory amount of variability in the scale. Based upon reliability analysis, all 13 items were retained for an overall Cronbachs alpha of .73 (Table 4-10). Table 4-10. Inter-item Consistency Statistics for the Internet Usage Construct (N=247) Usage Item Mean* SD Corrected Item total Correlation Alpha if item deleted How many hours a week do you spend on the Internet 2.1 .92 .53 .70 How many sites do you visit on an average session online 2.5 .91 .36 .71 Have you ever created a Website .20 .39 .28 .72 Download music 2.4 1.33 .33 .71 Read a blog 1.9 1.13 .37 .70

PAGE 89

75 Table 4-10 Continued. Inter-item Consistency Statistics for the Internet Usage Construct (N=247)* Usage Item Mean* SD Corrected Item total Correlation Alpha if item deleted Instant message 3.5 1.52 .36 .71 Read Facebook or MySpace 3.9 1.37 .25 .73 Watch videos 2.3 1.20 .44 .70 Shop online 2.4 1.10 .33 .71 Shop/sell on Ebay 1.7 .94 .30 .71 Use a search engine 4.3 .91 .43 .70 Work on WebCT or other online course 4.3 .90 .17 .73 How often do you use the Internet to find news/information online 4.0 1.10 .48 .70 *Five-point response scale where 1= very little to 5= very often. After the data was summated, respondents we re then categorized into a high or low Internet usage based on a means split of 2.71. The median for the group was 2.70 and a mode of 2.38. Based on the means split, 128 partic ipants were considered to be lower in their Internet usage and 119 were considered to be higher in their Internet usage (Table 411). Table 4-11. Means Table for Internet Usage. n (n%) Mean SD Low Internet Usage 128 (51.8%) 2.30 .30 High Internet Usage 119 (48.2%) 3.15 .32 Attitude Constructs Based upon previous research, three attit udinal scales were developed to assess general attitudes toward the In ternet and attitudes toward th e treatment or control version of the site to which they were exposed. Fo r the hypothesis testi ng, the scale measuring attitudes toward the treatment or control version of the site to which they were exposed was utilized. As discussed in Chapter 3, the most often used way to test attitude is through semantic differentials (Eagly & Chaiken, 1993). Two 11-item semantic-

PAGE 90

76 differential scales were used with bi-polar adjectives plac ed at the end of five-point scales. The scale measuring attitude toward the Internet in general was utilized for descriptive demographic reporting only and sh owed standard deviations from .8 to 1.1, indicating a satisfactory amount of variability. The coefficien t alpha reliability for the index was =. 70 (Table 4-12). The summated mean for the overall scale was 3.2 (SD= .91). Indicating a moderately positive att itude toward the Internet in general. Table 4-12. Inter-item Consistency Statistics for the Attitude Toward the Internet in General (N=249) Usage Item Mean* SD Corrected Item total Correlation Alpha if item deleted Good 3.5 .88 .40 .67 Credible 3.0 .80 .43 .67 Unbiased 2.4 .90 .20 .70 Easy to Understand 3.6 1.00 .22 .70 Important 3.1 1.00 .45 .66 Interactive 3.3 .96 .20 .70 Easy to Find 3.6 1.10 .21 .70 Beneficial 3.6 1.00 .47 .66 Believable 3.1 .76 .42 .67 Trustworthy 2.8 .77 .43 .67 Accurate 3.0 .77 .50 .66 *Five-point response scale where 1= very little to 5= very often. Attitude toward the treatment or control ve rsion of the site to which they were exposed was measured after exposure to the version of the site on an 11-item semantic differential scale. Based upon th e reliability analysis, 10 of the items were retained. The standard deviations for the scale ranged from .78 to 1.1 (Table 4-13). The coefficient alpha reliability score for the index was =.80. The summated mean for the overall scale was 3.88 (SD=.94).

PAGE 91

77 Table 4-13. Inter-item Consistency Statistics for the Attitude Toward the Treatment or Control Version of the Site to Which They were Exposed (N=237) Usage Item Mean* SD Corrected Item total Correlation Alpha if item deleted Good 4.0 1.0 .49 .74 Credible 3.7 .94 .45 .75 Unbiased 3.7 .93 .35 .76 Easy to Understand 4.2 .97 .41 .75 Important 3.8 1.0 .47 .74 Interactive 3.9 1.1 .32 .77 Easy to Find 3.9 .96 .54 .74 Beneficial 4.1 .88 .48 .74 Believable 3.8 .78 .56 .74 Trustworthy 3.7 .82 .56 .74 *Five-point response scale wher e 1=negative to 5= positive. The last scale which was also utilized for demographic description only was an eight-item index adapted from Rubin (1985) and Conway and Rubin (1991), where participants indicated their level of agreement along a fi ve-point scale of the importance of the Internet in thei r lives. A reported internal reliabil ity of .86 was reported in previous research (Ko, 2001). The standard deviations fo r the scale in this study ranged from .77 to 1.51. The coefficient alpha reliability score was =.72 (Table 4-14). A summated mean for the overall scale was 3.27 (SD= 1.02).

PAGE 92

78 Table 4-14. Inter-item Consistency Statistics for the Importance of the Internet in Subjects Lives (N=237) Usage Item Mean* SD Corrected Item total Correlation Alpha if item deleted I would rather surf the Internet than do something else. 2.63 1.02 .35 .70 My knowledge increases as my Internet usage increases. 3.02 1.05 .41 .69 It would be very difficult for me to survive without the Internet for several days. 3.10 1.51 .42 .70 Internet users are better educated people. 2.54 1.15 .45 .68 The Internet opens doors that would otherwise be closed. 4.03 .95 .46 .68 Information online should be engaging. 3.70 .77 .50 .68 Information online should be interactive. 3.54 .84 .40 .69 Information online should be entertaining. 3.63 .86 .37 .70 *Five-point response scale where 1=str ongly disagree to 5= strongly agree. Information Recall For the purpose of hypothesis testing, data on unaided recall was utilized. For the unaided recall portion of the st udy, participants were asked to list all of the information they could recall from the treatment or contro l version of the site to which they were exposed. The resulting qualitative data was cont ent analyzed and all true statements were scored as a +1 while untrue statements were coded as a -1 (Davis et al, 2005). Scores ranged from -2 to 16 (n=255). Two individuals indicated items that were not presented on the site, so their statements were coded ne gatively. A grand mean of 4.27 (SD=2.88) was calculated (Table 4-15).

PAGE 93

79 Table 4-15. Descriptive Report for Unaided Recall (N=255) N Minimum Maximum Mean SD Information Recall 255 -2.00 16.00 4.27 2.88 Hypotheses Tests Several hypotheses were made based on the independent and intervening effects of Internet usage, problem-solving style, and si te interactivity on subjects attitude and information recall. An overall means table (Tab le 4-16) provides insi ght into the average attitudes toward the treatment/control versi on of the site split by low/high level of Internet usage, adaptive/innovative problem-s olving, and the version of the site given. Problem-solving was grouped by adaptive a nd innovative rather than showing the variable as continuous to he lp with readability of the statistics (Graziano & Raulin, 2000). Table 4-16. Means for Attitude Toward Tr eatment/Control split by Low/High Level of Internet Usage, Adaptive/Innovative Probl em-solving Style, and Experimental Condition Presented (With Cell Sizes) Experimental Condition More Adaptive More Innovative Low Internet Usage High Internet Usage Low Internet Usage High Internet Usage Interactive 3.45 3.59 3.57 3.39 (36) (35) (22) (16) Non-Interactive 3.57 3.54 3.44 3.58 (32) (30) (29) (30) Total 3.50 3.57 3.50 3.51 (68) (65) (51) (46) An overall means table (Table 4-17) shows the average information recall split by low/high level of Internet usage, adaptiv e/innovative problem-sol ving, and the condition presented with.

PAGE 94

80 Table 4-17. Means for Information Recall spli t by Low/High Level of Internet Usage, Adaptive/Innovative Problem-solving Style, and Experimental Condition Presented (With Cell Sizes) Experimental Condition More Adaptive More Innovative Low Internet Usage High Internet Usage Low Internet Usage High Internet Usage Interactive 4.74 4.33 4.30 3.78 (38) (36) (23) (18) Non-Interactive 4.15 4.12 4.18 4.58 (34) (33) (33) (31) Total 4.46 4.23 4.23 4.29 (72) (69) (56) (49) H1: Unaided information recall and level of attitude toward an information-driven Extension Website will diffe r significantly as a function of site interactivity, problem-solving style, and Le vel of Internet usage. It was predicted that problem-solving styl e, level of Internet usage, and site interactivity will affect the attitude toward an information-driven Extension Website and the information recalled from that site. To test this, a multivariate analysis of variance (MANOVA) was run (Table 4-18). MANOVAs offe r a more sophisticated analysis of the variables allowing for the exploration of multiple independent and dependent variables (Graziano & Raulin, 2000). Results showed a partial support for this hypothesis. No significant three-way interac tion was found between problem -solving style, level of Internet usage, and site in teractivity on attitude (F=.67, p=.80) or information recall (F=1.50, p=.13). Results also indicated no tw o-way interactions between problem-solving style and site interactivity on attitude (F=.81, p=.75), probl em-solving style and Internet usage on attitude (F=.65, p=.92) or site interac tivity and Internet usage on attitude (F=.26, p=.61). However, significant two-way interactions were found between problemsolving style and site interactivity on info rmation recall (F=1.60, p=.05), problem-solving

PAGE 95

81 style and Internet usage on information reca ll (F=1.84, p=.01), and site interactivity and Internet usage on information recall (F=9.53, p=.00). Results indicated no main effects for pr oblem-solving style on attitude (F=.64, p=.97), site interactivity on at titude (F=.05, p=.83) and info rmation recall (F=.40, p= .53), Internet usage on attitude (F=.02, p=.89) and information recall (F=.21, p= .65). However, a significant main effect for pr oblem-solving style on information recall was found (F=2.12, p=.00). Table 4-18. MANOVA Results for Problem-solving Style, Site Interactivity, and Level of Internet Usage on Attitude a nd Information Recall (N=229) Source Df F P Problem-solving Style (PS) Recall 61 2.12 .00* Attitude 61 .64 .97 Level of Internet Usage (IU) Recall 1 .21 .65 Attitude 1 .02 .89 Site Interactivity (SI) Recall 1 .40 .53 Attitude 1 .05 .83 PS x SI Recall 31 1.60 .05* Attitude 31 .81 .75 PS x IU Recall 36 1.85 .01* Attitude 36 .65 .92 SI x IU Recall 1 9.53 .00* Attitude 1 .26 .61 PS x SI x IU Recall 14 1.50 .13 Attitude 14 .67 .80 Error Recall 80 (5.55) Attitude 80 (.31) Note. Values enclosed in parenthe ses represent mean square errors. Across the whole design, significant tw o-way interactions between problemsolving style and site interactivity were found for information recall, indicating the degree to which one is more adaptive/innova tive and the level of site interactivity affected how much information was recalled. A significant two-way in teraction was also found for problem-solving style and Internet us age for information recall, indicating that

PAGE 96

82 the degree to which one is more adaptive/innova tive and their level of Internet usage may have an effect. Means tables further dem onstrates the relationship between problemsolving style and site interactivity (Table 4-19) and problem-solvi ng style and Internet usage (Table 4-20). Table 4-19. Means for Level of Problem-s olving Style and Site Interactivity on Information Recall Overall Site Interactivity Problem-solving Mean N Std. Deviation Non-interactive Adaptive 4.13 67 2.97 Innovative 4.38 64 2.60 Interactive Adaptive 4.47 78 2.77 Innovative 4.00 46 3.31 Table 4-20. Means for Problem-solving Style and Internet Usage on Information Recall Overall Usage Problem-solving Mean N Std. Deviation Low Adaptive 4.46 72 3.08 Innovative 4.23 56 2.96 High Adaptive 4.23 69 2.65 Innovative 4.29 49 2.97 The significant two-way inter action between site interactiv ity and level of Internet usage on information recall suggests that in formation recall may differ based on level of Internet usage and site interactivity. A mean s table further demonstrates the relationship between site interactivity and Internet usage (Table 4-21). Table 4-21. Means for Site In teractivity and Internet Us age on Information Recall Site Interactivity Usage Mean N Std. Deviation Non-interactive Low 4.16 67 2.74 High 4.34 64 2.85 Interactive Low 4.57 61 3.30 High 4.15 54 2.70 The significant main effect shows that di fferences may lie in the level of problemsolving style. More adaptive individuals had a mean recall of 4.32 (SD=2.86) and more

PAGE 97

83 innovative individuals had a mean recall of 4.22 (SD=2.91) (Table 4-22), indicating that individuals who were more adaptive recalled information better overall. Table 4-22 Means for Problem-sol ving Style on Information Recall Problem-solving N Mean Std. Deviation Adaptive 141 4.32 2.86 Innovative 105 4.22 2.91 For Subjects who Received the Interactive Site H2: Attitude toward an information-driven Extension Website will differ significantly as a functio n of Internet usage and problem-solving style. It was expected that when holding site inte ractivity constant to include only those subjects who were exposed to the interactive version of the site, attitudes would differ as a function of Internet usage and problem-sol ving style. Results show no support for the hypothesis. The ANOVA results (Table 4-23 ) show no two-way interaction between problem-solving and Internet usage (F=.68, p=.84). No main effects were found for problem-solving (F=.62, p=.94) or Internet usage (F=.00, p=.97) on attitude toward an information-driven Extension Website. Table 4-23. ANOVA Results for Those View ing the Interactive Version, Problemsolving Style, and Internet Usage on Attitude Toward and Information-Driven Extension Website (N=109) Source df F P Internet Usage (IU) 1 .00 .97 Problem-solving Style (PS) 47 .62 .94 PS x IU 23 .68 .84 Error 37 (.34) Note. Values enclosed in parenthe ses represent mean square errors. H3: Unaided information recall will diffe r as a function of Internet usage and problem-solving style. It was expected that when holding site inte ractivity constant to include only those subjects who were exposed to the interactive version of the site information recall would

PAGE 98

84 differ as a function of Internet usage and pr oblem-solving style. Results show support for the hypothesis. The ANOVA results (Table 4-24) show a significant two-way interaction between problem-solving and Internet usag e (F=2.04, p=.02, cohens f = .59). Significant main effects were found for problem-solving (F=2.19, p=.00) and Inte rnet usage (F=9.77, p=.00) on information recall. Table 4-24. ANOVA Results for Those View ing the Interactive Version, Problemsolving Style, and Internet Us age on Information Recall (N=115) Source Df F P Internet Usage (IU) 1 9.77 .00* Problem-solving Style (PS) 49 2.19 .00* PS x IU 23 2.03 .02* Error 41 (5.25) Note. Values enclosed in parenthe ses represent mean square errors. For those receiving the interactive site, information recall differed significantly based on problem-solving style and Internet usage. A means table sheds light on the interaction between problem-solving and Internet usage (Table 4-25). Table 4-25. Means for Problem-solving Style and Internet Usage on Information Recall for Individuals Viewing the Inte ractive Version of the Site Usage Problem-solving Mean N Std. Deviation Low Adaptive 4.74 38 3.02 Innovative 4.30 23 3.78 High Adaptive 4.33 36 2.53 Innovative 3.78 18 3.06 A means table further describes the main ef fects of Internet usage (Table 4-26) and problem-solving styles (Table 4-27). Based on th e means table, subjects lower in Internet usage had slightly higher information recall than those higher in Internet usage.

PAGE 99

85 Table 4-26. Means for Internet Usage Main Effects on Information Recall for Those Viewing the Interactive Version of the Site Source Level Mean N Std. Deviation IU High 4.15 54 2.70 Low 4.57 61 3.30 Those who are more adaptive had higher info rmation recall than subjects who were more innovative. Table 4-27. Means for Problem-solving Style Main Effects on Information Recall for Those Viewing the Interac tive Version of the Site Source Level Mean N Std. Deviation PS Adaptive 4.47 78 2.78 Innovative 4.00 46 3.30 For Subjects Who Received the Non-Interactive Site H4: Attitude toward an information-driven Extension Website will differ significantly as a functio n of Internet usage and problem-solving style. It was expected that when holding site inte ractivity constant to include only those subjects who were exposed to the non-interactive version of the site, attitude would differ as a function of their level of Internet usage and problem-sol ving style. Results show no support for the hypothesis. The ANOVA resu lts (Table 4-28) show no two-way interaction between problem-so lving and Internet usage (F=. 50, p=.97). No main effects were found for problem-solving (F=.93, p=.60) or Internet usage (F=.40, p=.53) on attitude toward an informa tion-driven Extension Website. Table 4-28. ANOVA Results for Th ose Viewing the Non-Interact ive Version of the Site, Problem-solving Style, and Internet Usage on Attitude Toward and Information-Driven Extension Website (N=121) Source Df F P Internet Usage (IU) 1 .40 .53 Problem-solving Style (PS) 48 .93 .60 PS x IU 27 .50 .97 Error 44 (.29) Note. Values enclosed in parenthe ses represent mean square errors.

PAGE 100

86 H5: Unaided information recall will differ significantly as a function of Internet usage and problem-solving style. It was expected that when holding site inte ractivity constant to include only those subjects who were exposed to the non-interactive version of the site, unaided information recall would differ as a functi on of their level of Intern et usage and problem-solving style. Results show partial support for th e hypothesis. The ANOVA results (Table 4-29) show a significant two-way in teraction between problem-s olving and Internet usage (F=1.69, p=.05, Cohens f = .53). Significant main effects were found for problemsolving (F=1.65, p=.04), while no significant main effects were found for Internet usage (F=1.58, p=.22) on information recall. Table 4-29. ANOVA Results for Those Viewi ng the Non-Interactive Version, Problemsolving Style, and Internet Us age on Information Recall (N=115). Source Df F P Internet Usage (IU) 1 1.58 .22 Problem-solving Style (PS) 49 1.65 .04* PS x IU 31 1.69 .05* Error 49 (5.50) Note. Values enclosed in parenthe ses represent mean square errors. Results indicate that for t hose receiving the non-interactiv e site, information recall differs based on problem-solving style and Inte rnet usage. A means table sheds light on the interaction between problem -solving style and Internet us age (Table 4-30). For those viewing the non-interactive site, innovators are higher in their information recall regardless of their level of Internet usage.

PAGE 101

87 Table 4-30. Means for Problem-solving Style and Internet Usage on Information Recall for Individuals Viewing the Non-In teractive Version of the Site. Usage Problemsolving Mean N Std. Deviation Low Adaptive 4.15 34 3.16 Innovative 4.18 33 2.28 High Adaptive 4.12 33 2.80 Innovative 4.58 31 2.92 A means table further describes the signi ficant main effect for problem-solving style (Table 4-31). Based on the means table, more adaptive subjects who received the non-interactive site were slightly lower in their recall than more innovative subjects. Table 4-31. Means Problem-solving Main E ffects on Information Recall for Those Viewing the Non-Interactive Version of the Site. Source Level Mean N Std. Deviation PS Adaptive 4.13 67 2.97 Innovative 4.38 64 2.60

PAGE 102

88 CHAPTER 5 DISCUSSION Overview Research suggests that indi viduals are increasingly turn ing to the Internet when seeking out information on agri cultural topics that have long been the subjects of traditional Extension programming, such as food safety, economics, biotechnology, consumer sciences, and horticulture. Yet li ttle is known about how characteristics of individuals and the online environment itself might interact to affect processing of information within the context of Extension. This study examined the effects of pr oblem-solving style, level of Website interactivity, and Internet us age on the attitudes of a samp le of young adults toward an information-driven Extension Website and recall of the information presented on that site. Extension and other agri cultural sources are increasingly using the Internet to disseminate scientific information. However, while this information is important to audience members lives, its complex nature makes it difficult to present, especially in an online environment, where information-seeker s may vary significantly in how they attend to, process, and perceive what is being communicated. Limited research exists as to how traditional and non-traditional us ers may respond to Extensions efforts to employ more Web-based channels of communication for nonformal education purposes. Further, few studies to date have looked at how young adults, who combine relatively limited experience of Extension with relatively high usage of the Web, process Web-based Extension information when at tempting to solve a problem.

PAGE 103

89 Results of this study suggest that for these audiences, the cognitive style with which individuals solve problems, combined wi th their level of usag e of the Internet, do affect the level of information that they r ecall. Based on the results of this study, which showed significant effects in the area of unaided recall of information, characteristics such as problem-solving style may represen t important factors fo r Extension educators and communicators to take into considera tion when developing content and designing effective Web-based information. During the in formation-seeking process, individuals problem-solving styles may affect how they lo ok at and experience information online. It is therefore important that researchers take inventory of what methods of information presentation online influence different problem-s olving styles. It is al so important to take inventory of how young adult audiences view this information. These audiences, while not traditional Extension clientele, represent po tential future users of Extension as well as many of the main current users of online information. To carry out this investigation, a between -subjects factorial design was utilized to conduct a randomized post-test only experiment with a sample of young adults who were directly administered the KAI problem-solving style assessment inventory and then asked to report their level of Inte rnet usage, general attitude toward the Internet and demographic information through a series of questions. Respondents were then randomly assigned to one of two versions of an Extension informationdriven Website that focused on the economic aspects of car buying, a problem -solving situation that is salient with young adults. The versions did not differ in messag e, but in level of in teractivity available to the user. The level of in teractivity was defined by one site offering no opportunities to vary the experience with the site, and the other version a llowing for several paths through

PAGE 104

90 the site and moving objects with which the user could interact. Respondents were then asked through an online survey to report a ttitudes toward the site version viewed, previous experiences with Extension, and the topic represented on the site, and to indicate information recalled. A panel of experts wa s utilized to look at validity of the instruments. Afterwards a p ilot study was conducted with a similar population to not only test the reliability of the in struments, but also to conduct a manipulation check on the two versions of the site. Data analysis was conducted and results we re presented previously in Chapter 4. A total of 256 students who were en rolled in two College of Ag ricultural and Life Sciences courses at a large Southeastern university participated in th e study. Most respondents were female (n=145, 56.9%), with 110 (43.1 %) being male. The majority of respondents were 18-20 years old (56.3%) followed by respondents 21-23 years old (37.9%), respondents 24-27 years old (5.1%), and res pondents 28 years or older (0.8%). This chapter will present the key findings, imp lications, limitations, recommendations for theory and practice, and conclusions to the study. Key Findings The demographic results indicated that th e majority of participants (98.8%) own a computer. While at home, the majority conne ct to the Internet using high-speed (55.3%) and wireless access (37.5%). At school, almost half (49.8%) of the respondents indicated using a computer lab to go online. The subjects preferred use of the Internet (60.9%) to any other media to seek out news and info rmation. While many did not have a Webpage or a blog, 85.2% did indicate using Websites like Facebook. These demographic findings support the literature th at this age group is a heavy user of the Internet for information and entertainment (Eastin, 2001; Fallows, 2004). Interestingly, while respondents

PAGE 105

91 indicated the Internet was easy to understand, important, beneficial, believable, and accurate, their overall mean for these items was only slightly positive, indicating that while the Internet is a tool used in their ev eryday lives, these subjects were still cognizant that not everything presented to them is necessarily accurate or unbiased. Subjects were also asked to describe thei r usage of Extension. Previous Extension research has indicated a need to seek out ne w audiences (Bull et al., 2004). This study has attempted to look at one such audience, young adults. Subjects reported being fairly low users of Extension services and information, im plying that this is an audience that has not been traditionally reached with curr ent Extension programs and messages. A total of five major hypotheses were test ed in this study. Hypothesis 1 postulated that: Unaided information recall and level of attitude towa rd an information-driven Extension Website will differ significantly as a function of site interactivity, problemsolving style, and level of Internet usage Hypothesis 1 was partially supported. Findings indicated no significant threeway interaction between probl em-solving style, Internet usage, and site intera ctivity on attitude and recall. No two-way interactions were found between problem-solving style and site intera ctivity, problem-solvi ng style and Internet usage, and Internet usage and site interactiv ity on attitude. Howeve r, findings indicated there were three significant two-way inter actions: 1) between pr oblem-solving style and site interactivity; 2) between problem-solving style and Internet usage; and 3) between site interactivity and Internet usage, all on information recall. While no main effects were found for site interactivity or Internet usage, main effect s were found for problem-solving style on information recall. Find ings from means tables sugge sted that subjects who were more adaptive in their problem-solving style and who received the in teractive version of

PAGE 106

92 the site were better able to recall information than those w ho were more innovative, while those who were more innovative in their pr oblem-solving style recalled information better if they received the non-in teractive version of the site. In hypotheses 2-5, site interactivity was he ld constant in order to explore the influences of the other major independent vari ables. Research indicates that interactivity may or may not have an influence on indi viduals attitudes an d information recall (Sundar et al., 2003; Sicilia et al., 2005; Teo et al., 2003; Li u, 2003). Results of this study supported one side of the literatu re base, in that differing levels of interactivity, in and of themselves, did not have an influenc e on attitude and information recall. Hypothesis 2 focused only on subjects receiv ing the interactive ve rsion of the site. It was postulated that attitude toward an information-dri ven Extension Website will differ significantly as a function of Inte rnet usage and problem-solving style. Hypothesis 2 was not supported. Problem solving style and level of Internet usage did not have a significant two-way interaction on attitude. Furthermore, no main effects were found for Internet usage or problem-solving style on attitude. Hypothesis 3 also focused on subjects receiving the interactive version. According to hypothesis 3, unaided information recall will differ as a function of Internet usage and problem-solving style Hypothesis 3 was supported. A si gnificant two-way interaction was found between problem-solvi ng style and Internet usage on information recall. Based on means tables, it is suggested that those subjects who were more adaptive and higher in their level of Internet usage also tended to be higher in their information recall. Those who were more adaptive and lower in their In ternet usage also tended to be higher in information recall than those who were more innovative. Significant main effects were

PAGE 107

93 also found for problem-solving style and Intern et usage on information recall. Individuals viewing the interactive site who were low in Internet usage had higher mean recall. Hypothesis 4 focused only on subjects rece iving the non-interac tive version of the site. It was postulated that attitude toward an informationdriven Extensio n Website will differ as a function of Internet usage and problem-solving style. Based on the findings, hypothesis 4 was not supported. For individua ls who viewed the non-interactive site, problem-solving style and level of Internet usage did not have a significant two-way interaction on attitude. Furthermore, no main effects were found for Internet usage or problem-solving style for attitude. Hypothesis 5 focused only on subjects rece iving the non-interac tive version of the site. It was suggested that unaided information recall will diff er as a function of Internet usage and problem-solving style. Hypothesis 5 was partially supported. A significant two-way interaction was found between probl em-solving style and Internet usage on information recall. Means tables suggested that those subjec ts who were more innovative and higher in their level of Inte rnet usage tended to be higher in their information recall. Those who were more innovative and lower in their Internet usage also tended to be higher in information recall than those who were more innovative. While no main effect was found for Internet usage, a main eff ect for problem-solving style on information recall was found. It appears that for indivi duals viewing the n on-interactive site, individuals who were more innovative in their problem-s olving style recalled more information than those who were more adaptive.

PAGE 108

94 While findings for hypotheses 2-5 were similar, it is interesting to note that for the non-interactive version of the website, Internet usage influenced reca ll as a main effect, while there was not a main effect for users w ho viewed the interactiv e version of the site. Implications of the Study The findings from this study suggest severa l significant theore tical and practical implications. While researchers looking at othe r cognitive process attributes, like need for cognition, have found that individuals exposed to interactive sites processed information more thoroughly than those exposed to non-in teractive sites (Sicilia, Ruiz, & Munuera, 2005), this study found that when problem-solvi ng and level of Internet usage are added to the equation, differences in terms of information recall emerge. These findings not only support previous findings in the usage and gratifications literatu re, but they also add to the literature base in agricultural comm unications by introducing this new component of problem-solving in the cont ext of Extension information as disseminated via the Web. The next section presents implications of the study accordi ng to problem-solving style and Internet usage ba sed on site interactivity. Problem-Solving Style Results of the study suggest on initial inspection that individuals who are more adaptive in their problem-solv ing style may be better able to attend to information regardless of the level of site interactivity, as opposed to those who are more innovative. However, in this study, individuals who are more innovative in their problem-solving style but were exposed to the non-interactive version of the site had a higher recall of information than those who were adaptive in their problem-solving style. These findings are supported by the literature in Adaption/ Innovation theory, which states that more adaptive types tend to think lin early and work through existing structure, while more

PAGE 109

95 innovative types may be more adventurous and will work non-linearly, using trial and error to work through problems (Pershyn, 1994; Foxall & Bhate, 1993). Based on the above, it may be that more innovative types te nd to work through an interactive Website in a more non-linear way than those who are more adaptively inclined, and may, therefore, attend to many di fferent things, not just focusing on the main components. Consequently, they may be less likely to reca ll the message presented. While research has shown that interactivity is us eful in drawing audiences to a site and keeping them there (Chen and Yen, 2004; Cho, 2003), the findings of this study indicate that interactivity may sometimes impede the recall of informati on presented on a site for specific problemsolving styles. For information-oriented sites, high levels of inter activity may therefore not be as beneficial as vary ing the level of interactivity, if the objective is recall of information. Based on the results of this st udy, if designers are in a position where they can vary site interactivity, they should consider doing so, by possibly offering an interactive entry page or interactive pages that do not contain the important information. Pages containing important or complex info rmation should be placed on less interactive pages to accommodate differi ng problem-solving styles. It could also be the case, based on the findings of the study that innovative types may be more difficult to reach with message s. The Adaption-Innovation theory discusses the fact that individu als who are presented with a situa tion in which they cannot utilize their preferred problem solving style will revert to a coping behavior in which they adjust themselves to the situation (Kirton, 1989). In this process they will need to bridge the gap between what they prefer and what th ey perceive as necessa ry in the situation. Individuals will do this as long as is deemed necessary or as long as they can tolerate it

PAGE 110

96 (Kirton, 1989). This coping behavior will not change ones problem solving style, but may cause an individual to want to leav e the situation sooner. In this study, the innovators who entered the situ ation in which they were fo rced to work linearly through the site may have been working through a coping behavior in which they wanted to complete the task quickly. Thus, they could have attended to the message more because they got into the site, found what they needed and left, whereas those who were presented with the interactive version di d not go into a coping behavior and stayed on the site playing and not focusing on the task. Adapti on/Innovation theory stat es that innovators, when allowed to work within their preferred style, will generate many ideas, but may not find themselves at a conclusion or end to a problem (Pershyn, 1994). Those who were allowed to work in the interactive site may th erefore have been able to work within their style and not come to a conclusion on the topic. Based on this studys results, it may be the case that recall is not only affected by an individuals problem-solving style, but al so may be a product of level of Internet usage. While Internet usage is considered to be generally high for this population, even for young adults, usage does vary. It was f ound in this study that individuals who are lower in Internet usage and who are more ad aptive in problem-so lving style recalled more information than indivi duals who are more innovative and lower in their Internet usage. Even for this young adult audience, t hose who are adaptive a nd lower users of the Internet may be better able to recall the information presen ted online than innovators who happen to be lower in their In ternet usage. Other venues may be better for reaching those innovative lower users; this study points to th e fact that other tec hnologies like ipods or more traditional methods like radio, print, or television need to be taken into account as

PAGE 111

97 well. No one medium will be as effective as a mix of relevant media when attempting to reach an audience via mass media channels. It was also found that thos e receiving the interactive ve rsion of the site who were also more adaptive in their problem-solving style had higher recall than those who were more innovative. As stated earlier, this fi nding could be because more adaptive types were possibly coping with the interactivity and low structure of the page and were working quickly to understand the informati on, while those who were more innovative may have been more comfortable and were not as attentive to the information presented on the site. Information recall literature cont ends that the Internet is helpful in the increased elaboration of information, but it also decreases information learning due to selective scanning by partic ipants (Eveland & Dunwoody, 2002). Based on the definition of innovators as being more reckless in th eir information seeking, it could be concluded that subjects in this study were more likely to participate in this selective scanning and thus recall less. It could be inferred that Extension communicators trying to appeal to more adaptive types would be safe in presenti ng information in an interactive format, as these users will be able to push through the inter activity to get to the context. However, if they are attempting to appeal to more innova tive types, communicators must be cognizant of the level of interactivity on the site, as it could impede information recall. When trying to reach a variety of problem solving styl es, one site that takes both ends of the continuum into account is needed. Based on Kirtons work by understanding th e make-up of an audience, such as their gender or profession, a designer can have a good idea of what problem-solving style may prevail in that audience. For example, Ki rton has reported that people in the teaching

PAGE 112

98 profession have shown KAI means of 95.0-97.0, and marketing managers in the United States had a mean of 102.2 (Kirton, 1999) indicating that by understanding the demographics and occupations of your audi ence you could be able to infer the most prevalent problem-solving style of the group. By taking such information into consideration, a site can be de signed with more structure that leads users through the site, rather than allowing them many disparat e paths through the important information. Designs for more innovative audiences could al so include the information of most value on one page so they are exposed to the w hole message and do not miss things as they would if the information was broken out thr ough hyperlinking or inte ractivity. However, care must be taken as to keep the attention of these innovative audi ences who may prefer to work in the non-structured environment. Internet Usage Results of this study also indicated th at for individuals exposed to the noninteractive version of the site, those who were higher in Intern et usage recalled more than those who were lower in their usage. This could be because those who are higher in usage are used to gathering information online from information-laden sites than less frequent users and thus may be better able to process information and recall it more. The theory of Uses and Gratifications states that indi viduals choose a medium to gratify their information needs based on memories of past media usage (Katz, Blumler, & Gurevitch, 1974). This previous knowledge of the media will drive their future usage and experiences with the medium in question (Katz, Blumler, & Gurevitch, 1974). Thus, those who have used the Internet more will have previous expecta tions and experiences that will tend to drive their processing, cau sing them to possibly gr atify their need of finding information more successfully than users not as familiar with the medium.

PAGE 113

99 For Extension communicators who are design ing sites with interactivity, it is crucially important that they have a basic unde rstanding of the level of Internet usage for the target audience to ensure effectivene ss. While this would be a daunting task, by understanding the demographics of an a udience through basic audience analysis techniques, designers can then develop sites to match their us ers level of usage. While the adoption of the Internet by users and Ex tension communicators continues to increase, new technologies and advancements will contin ue to emerge, indicating that there will always be differing levels of usage of technol ogy that will affect user s viewing of sites. Thus, level of usage is something that mu st be continuously monitored for each new Website. Based on Uses and Gratifications th eory, initial experiences with a Website must be gratifying and successful for users, if they are to return. With each new technology experience users are faced with a different way to gratify needs. Thus young adult users who are experiencing the Extension sites for the first time will need to have good first experiences. Interestingly, in this study, attitude was not a factor. Problem-solving style, Internet usage, and level of site interactivity did not influence the attitude of individuals on the site presented in this instance. While past research has shown that interactivity can influence positive attitudes toward political candidates and advertisements (Teo et al., 2003; Sundar et al., 2003; Chung & Zhao, 2004), this study presented the same information on both versions of the site, vary ing only site interac tivity. The non-effect on attitude could be explained by the fact that younger audiences, like the one being studied, also tend to be higher users of the Internet and have been engaged with the medium longer than adult users (Eastin, 2001; Fallows, 2004). It could be assumed that due to

PAGE 114

100 their high use of the medium, subjects attit udes were not affected by the level of the interactivity, due to expectat ions from previous usage. Studies show that there are a variety of sites that provide information in highly interactive formats while others provide similar information using non-interactive methods (Henkia, 1990; Amant, 2005). Young adults are exposed to the Internet on a daily basis, thus possibly erasing its novelty in their minds. This audience is accustomed to looking at a variety of sites that co ntain both interactiv e and non-interactive information, which could mean that some of these users may screen out the varying interactivity levels and focus onl y on the information consciously. In contrast, it was implied through the fi ndings that those receiving the interactive version of the site but who were lower in us age had higher recall. One explanation of this could tie back to the uses and gratificati on literature base. Lower users who do not have previous expectations of the media (Katz, Blumler, & Gurevitch, 1974) may thus have been more attentive to the interactivity a nd the information presented on the site. The novelty of this new technology for these user s could make them attend to the message more. The attention getting and keeping nature of this interactive format may engage these audiences and keep them moving th rough the information in a more efficient manner. Limitations This study was exploratory in nature, and can shed light on future research in the areas of problem-solving style, informati on processing, and usage of the Internet by Extension audiences. However, some limita tions do exist that should be noted. As discussed in the literat ure review, interactivity is defined and described differently by different theorists, and even us ers have different ideas of what interactivity

PAGE 115

101 is (Liu, 2003). This study used two different levels of interactivity as the treatment and control conditions, based on the definiti on of interactivity given by Sundar and colleagues (2003) in which the number of hype rlinks or paths through the information were equated with increased levels of intera ctivity. For this study th e interactiv e version of the site allowed for individuals to click th rough the information in any order, to move and close boxes which popped-up, and interact with a car graphic th at moved across the screen. The non-interactive version was a pa ge containing only text and one static graphic. Other definitions of interactivity have included the idea that interactivity is on a continuum that differs based on the amount of control the user has and his or her increased sense of place (Downes & McMill an, 2000). Jensen (1998) described four types of communication in interactivity, wh ich includes registra tion (two-way with feedback based on what the user inputs ), continuous (two-way communication), consultation (pre-defined content a user seek s out with a feedback loop), and allocution (one-way communication with no feedback). This study is thus limited in the fact that only one definition of interactivity was utilize d. Explorations of thes e different types of interactivity represent a direction for future re search. Varying levels of interactivity could also be utilized. This is an area in which further research is warranted. Another potential limitation to the study was the 48 subjects who were excluded from the sample due to the KAIs suspect sc oring considerations. The KAI was originally developed to be used with working adults in a business setting, but it has been deemed acceptable for use with populations as young as 13 (Kirton, 1999). While the KAI is a robust and reliable instrument, it has most ofte n been administered in small groups where the participants are comfortable and do not feel rushed (Kirton, 1999). In this study, the

PAGE 116

102 instrument was directly administered to the en tire sample of students in their course with the instructor and administrator walking around the room, a condition which could have led to interruptions in thought, time pressure, and surr oundings that have been found to possibly affect the wastage rate of the instrument (Kirton, 1 999). Kirton has stated that for a young population, such as the one used in this study, it is not uncommon to report wastage rates between 10% and 20% due to ma turity levels, and the students need to deliberately score in an acceptable way (Kirton, 1999, p.19). While care was taken by the administrator of the KAI to address these concerns, a 15% wastage rate was found in the responses. Based on the theoretical literature, this was deemed acceptable. Another limitation to this study is that it only measur ed immediate unaided recall and did not address recall over time. Some resear ch has indicated that time is needed after exposure to information to increase recall as people must process the information (Wicks, 1995). Based on the anonymity of the particip ants in the study, it was impossible to contact them for further recall testing and so only immediate unaided recall was measured. Another limitation to consider is that while the instrumentation included an established psychometric test of problem-s olving style, it al so included several researcher-developed items. This study was exploratory in nature and all items were created based on the literature To address this potential issue a panel of experts was utilized to look at the validity of the instrume nts, and data analysis was run to confirm the reliability. The resulting statistics showed re liabilities for the scaled items ranging from .70 to .80. Although these reliabilities were deemed acceptable for the purpose of this exploratory study, further testi ng on these items is warranted.

PAGE 117

103 A fifth limitation to account for is the us e of a college student sample. Since young adults were of specific consideration in this study as an audience, an audience which is considered to be both active in their usage of the Internet and underserved by traditional Extension communications, colleg e students were viewed as an appropriate population to study. A final limitation to this study is the admini stration of the site conditions and final instrument online. This method of testing is still relatively new in comparison to traditional experimental techniques. However, research has indicated that using the Internet to collect data is reliable and efficient. Dillman (2000) discussed the usage of Internet-based surveys as having the potential to bring great effici encies in design and administration to traditiona l surveying methods. Lander, Wingenbach, and Raven (2002) found that Web-based survey methods were as reliable for collecting social science data as were paper methods. By running this experime nt online, participants were able to look at the condition in a more natural setting, thus minimizing some of the internal reliability issues seen in many experimental conditi ons. Online administration is common for audiences of college students when st udying Websites and interactivity (Sundar, Kalyanaraman, & Brown, 2003). Recommendations for Theory and Practice Recommendations for Practitioners By understanding how problem-solving styl es affect users perceptions of information-driven Websites with respect to such attributes as at titude and recall of information, Extension, agricultural comm unicators, and commodity groups who are utilizing the Internet to reach audiences with science-based information will be better able to make use of communications proce sses to inform audiences, educate them, and

PAGE 118

104 affect productive change. If Extension utilizes technique s to understand not only the demographic make-up of their target audien ces, but also the psychological make-up of these constituents, they can better develop sites geared to succe ssful communication and education. The findings of this study offer new chal lenges to designers of online information. The push to have interactivity to draw in audiences and keep them on a site must be balanced with the best method in providing information. As shown through the results of this study, it may be the case that communicat ors attempting to provide information to help users solve a problem may need to provi de content and design el ements that appeal to a diverse range of problem-so lving styles in one package. For audiences that tend to be more innovative in their problem-solving styles less-interactive forms of media may help in recall of information as they will po ssibly work through the information more efficiently, while more adaptive types will be able to recall information regardless of level. Communicators wishing to keep audiences on their site through interactivity and yet elicit comprehension and r ecall of that information may ha ve to find a way to design a single Website in a fashion that forces more innovative problem-solving types of users to work with more structure, yet keeps their attention. A site deve loped to provide car buying or financial information, such as the one used in this study, must contain features that are attractive and draw attention, but as the user gets into the important parts of the information, it needs to be presented in a way that allows for more structure and linearity. This could be done not only through basic pa ges, but through direct feedback loops or online self-practice quizzes that direct the user in a linea r fashion through the important

PAGE 119

105 information. Extension communicators who are trying to reach broad audiences containing both innovative and adaptive types wi ll have to work carefully to ensure sites include successful components for both gr oups. This studys findings suggest that the structure of the site is as important as th e individual elements, based on the individual problem-solving styles of the target audien ce, and communicators must thus gauge what level of structure is n eeded in each situation. This outcome adds more challenges to designers, who may have learned that interactivity brings people into the site and keeps them there longer. It can be inferred as a result of this studys findi ngs that while interactivity can be beneficial; it may not always be beneficial in mass qua ntities. Designers cannot just a dd static elements to a site without any interactivity, because while us ers may attend to the message better, more innovative types might not enter the site or stay if it is not appealing to them. If they are forced to work in a coping behavior too long they may be more likely to not return to the site. By utilizing splash entry pages and in teractive home pages, and then having inner pages with minimal interactivity, designe rs may be able to reach both groups successfully. While it may be hard for designers and pract itioners to clearly define their audience in terms of problem solving style, it can be inferred for specific audiences based on the theoretical research that ha s been completed. For example, a study of bankers showed a mean KAI score of 91.3, managers had mean s ranging from 95 to 102.2, teachers ranged from 95 to 101.4 in studies done in the Unite d states, advertising professionals and designers showed a mean of 101, and poli ce had a mean score of 98.4 (Kirton, 1999).

PAGE 120

106 These professional breakdowns could help gui de designers when designing for specific audiences. It is important that we not only understa nd who we are communicating to, but also, who is doing the communicating. Today, many Webs ite designers are not just educated in graphic design, as print designers of the past often were. More Website designers today are versed in information technology and the many high levels of st ructure involved in technology. Past research by Kirton shows that in technical professions, like Website design, individuals tend to be more adaptiv e (Kirton, 1999). Thus, this difference also makes it important for designers to be aware of their own problem-solving styles when designing sites. Designers will tend not only to design to client specifications, but will also design things that are appealing to th em; thus, designers own problem-solving styles may affect what they believe to be an effec tive site. Individuals will want to avoid coping behavior if possible and work w ithin their preferred style, and as they design a site this could possibly carry over into what the site looks like. By being aware of their own problem-solving style and that of their audien ce, designers will better be able to match the design to be effective for all. As Extension moves to develop more online communities, it is important that communicators take inventory of those a udiences which they are trying to reach. Designers and developers need to keep in mind that when developing information for young adults, the Web may not be the only me thod that should be utilized. Based on individuals problem-solvi ng styles, other communication technologies such as television, print, DVDs, or cellular phones may be better suited. It is also important to understand that designers cannot just solve this issue by providing plain text sites with no

PAGE 121

107 attention-getting draw. There are many ways to design structure, even through the more complex programs like Flash, that draw a user in a fairly linear fashion through a site without them getting distracted, such as videos, quizzes, or feedback loops. As younger audiences who are accustomed to using the Internet mature and become more of a focus for Extension, communicators need to consider that their previous experiences with the medium may be fueling th eir current expectations and attitudes. It was found in this study that while such us ers may have neutral attitudes toward information online, they do not have highly pos itive attitudes toward the credibility and believability of specific information, such as th at presented in the si te stimulus. As shown by the findings of this study, these attitude s did not significantly change based on the level of interactivity present on the site. The Uses and Gratifications literature adds to this discussion in explaining that users will make choices as to which information to look at based on past experiences. These past experi ences have formed users attitudes toward the Internet. For audiences who are less experienced onlin e interactivity diffused in a site may impede individuals who are more innovative in their problem-solving style, causing them to take in less of the information pres ented. As communicators attempt to provide information to low Internet-using audiences, it may be necessary to provide low levels of interactivity until the audience is more versed in the medium. Extension and agricultural communicators must be concerned and cognizant of the level of Internet usage by their audiences when planning a new site. Agai n, while interactivity attracts audiences, designers may overdo it by letting the interactiv ity get in the way of users ability to use the information.

PAGE 122

108 Educators attempting to teach future comm unicators must address these issues, as well, in the classroom. Students must be made aware of the different types of psychological factors affecting the way i ndividuals look at the communication they produce. They need to be aware of the in fluence audience attri butes beyond simple demographics can have, and will need to learn how to assess that information in order to better communicate. It is importa nt that individuals who will be working in agricultural communications are aware that the way they present information-driven content may affect the amount the user retains. Students who are in agricultural communication programs are among the population who is most well-vers ed in the Internet. As show n through the literature and this study, their experiences online will form th eir attitudes. Students must not let their attitudes form their opinion on whether to use interactivity; this decision must be based on information as to the many positive and negative aspects of including interactivity in information-driven sites. While the findings of this study could offer new challenges to designers and communicators of Extension and agricultural information, it offers insight into how we can better reach audiences. As shown it is important that communi cators truly understand the audiences they are trying to reach. A udience analysis should be conducted by communicators that not only include de mographics, but also psychographics. Communicators can be more efficient in their profession if they take into consideration the effects of psychological factors like probl em solving style as well as factors like medium usage. Assumptions cannot be made on what is best for an audience based on

PAGE 123

109 past experiences. Technology is changing a nd peoples individual preferences are in constant flux, so communicators must continue to monitor these items. Future Research The findings of this study have the potent ial to inform research in the area of problem-solving and site interactivity. As science continues to make headlines, people will continue to seek out information from sources like Extension and agricultural communicators. Researchers should ensure that audiences continue to be assessed to discover the best ways to present informati on in ways that address various psychological differences, including problem-solving style. Repetition of this study is recommended to validate the findings. Much communication research has focused on the implications of the convergence of new technologies like radi o and television into societ y. As new communication venues emerge, such as the Internet, new studies mu st be conducted to desc ribe the new nuances and characteristics of that medium. Th is study offers new implications on how researchers can look at these emerging t echnologies. This study indicated that by combining the usage of the new medium w ith a psychological factor, like problem solving style, new patterns of how users use the medium can be discovered. Future studies of new technologies like RSS, ipods and cellular phones should be conducted to see if these same factors make a difference in the usage of the medium, attitudes toward the medium and information presented, and information recall. Due to the continuing argument for and against the impacts of interactivity on a site, further research should replicate this study with various levels of interactivity. By including more structure like: two-way interactions such as quizzes, discussions, and input fields, actions which allow the user to form the path in which they move through

PAGE 124

110 the information by links, or multimedia, further analysis could further define what types of interactivity are best suited for different levels of problem-solvi ng style and level of Internet usage. Future studies could co mpare interactivity based on animations, multimedia components like video or audio, or hyperlinking to look at the linearity structure used and preferred by different type s of problem solvers. It would also be beneficial to look at how these preferences differ based on different ages of audiences. More levels of interactivity could be employed in future studies to see if audiences would be better able to decipher between non-interactivity and interactivity. It would be beneficial to take a deeper inventory of what types of Web design elements, and the balance of content and in teractivity being utilized by Extension and agricultural communicators, aff ects user attitude and recall. Do audiences react better when certain levels of interactivity are in cluded? Do they recall information or have different attitudes based on the balance of cont ent and interactivity? Or, are they affected by basic graphic design elements utili zed, or not utilize d, by Extension? This study only assessed one component of Website design. It is important that when communicators are working with informat ion-rich sites for agriculture, that can also be tied to rich visuals, that they assess other components of Website design with younger as well as more traditional Extension a udiences. Do certain layouts, navigational schemes, colors, multimedia components, or f eedback features help in the use of these sites? Do individuals feel more satisfaction with any of these sites based on the navigation, colors, technology components, or feedback features? Do any of these design elements affect credibility of agricultural information for agriculture and non-agriculture audiences? Does the level of structure come into play with audiences looking at these

PAGE 125

111 sites? What are agriculture and Extension usi ng to reach these audien ces in terms of the various design components? Are they following design principles, and does that have an affect on the recall of information on these information-rich sites? This study only assessed the attitude toward the site version viewed in general. It is recommended that future studies look at at titude toward the information presented and attitude toward the technology after being exposed to the experimental version. This could be the missing link in which attit ude would differ based on usage and problemsolving style. The topic used in this study was shown to be of interest and relevance to this audience. Respondents indicated minimal expe rience with the topic of car purchasing. Future studies might look at a more complex, no vel, or controversial topic in agricultural science to see if processing pa tterns differ from what was seen in this study. It could be argued that car buying by nature has a structured path in which to move through the information, and another topic like biotechnology or food security may offer new insight into these findings. While connections were found between problem solving style, Internet usage, and information recall it may be beneficial to de lve deeper into the problem solving style of individuals to see what compone nts of that style truly made a difference. KAI is broken into three subscales: sufficien cy of originality, efficiency, and rule group conformity (Kirton, 1989). Sufficiency of originality, wh ich deals with the idea generation, research suggests that innovators may suggest ideas th at are both adaptive and innovative until they are sure of the boundaries of the stru ctures they are working in (Kirton, 1999). While it has been noted that these scales ma y not be as reliable w ith younger audiences,

PAGE 126

112 further analysis should explore the influen ces of these areas on information recall and attitudes. This further analys is may add more explanation to the findings of this study. While young adult populations are an a udience considered underserved by Extension, it is suggested to continue to rese arch other individual popul ations to see if the findings of this study are similar for other au diences. The mission of Extension is to bring research and information from the land grant university to the mass public. To reach this audience researchers must continue to monitor the best ways to present this information to help facilitate the learning and recall of important agricultural and scientific information. It may also be beneficial to follow through by comparing these different audiences to look for trends and similarities among generations. Based on the findings of Extension use by this population, it is recommended that future studies look more closely at younge r audiences use and know ledge of Extension. If young adults are an audience that may be future consumers of Extension information, researchers and communicators need to understand where they are getting information on topics like the one used in this study. It is also important to discover new ways to reach these audiences with this information. While this study looked at what method helped them in terms of recall, re searchers could also assess wh at method these constituents prefer. Other data collected in this study, incl uding qualitative data about individual definitions of interactivity, and motivations with the topic presented, should be analyzed for further understanding of th e subjects experiences with th e sites information and its impact on their informati on recall and attitudes.

PAGE 127

113 Designers of this information play an impor tant role in the final component that is published online. It would be helpful for researchers to take inventory of what agricultural and Extension comm unicators think their audiences need in an informationdriven site, and how that compares to what the audience actually needs. By understanding where knowledge gaps may exist, instructors can further the academic curriculum of these individuals. It may also be beneficial to do further testing with the designers of these sites to determine just how much thei r individual problem-solvi ng styles affect the structure of the sites they design and/or the levels of creativity th ey employ on the sites they develop. This research could also extend into more trad itional outlets of communication, such as how print design a nd exhibits are deve loped and received. Further testing should also address the issue of extended recall. While it is important to understand what helps individuals recall information directly after exposure, it is also helpful to see if one method over another will help in the remembrance of that information over extended periods of time. This audience of young adults may also just be a difficult audience to reach with technology. They are savvy and experienced in their usage, and due to that fact, other factors may be coming into play that affect th eir level of information recall and attitudes. Further research needs to furthe r explore this age of users to discover what could also be affecting the way they use and recall inform ation presented through newer technologies. Researchers may want to c ontinue to look at the re lationship between problemsolving style and Uses and Gratifications theo ry. Examination further into gratifications

PAGE 128

114 received through various media by psychological factors, such as problem-solving style, may add to the deeper literature base. Conclusions As Extension moves to designing inform ation online to reach audiences with information-rich topics, such as becoming more informed on scientific news, it is beneficial that this information be presente d in a form that will be usable, valuable, appropriate, and easy to recall. The findings of this study add to the theo ry base by tying together two very well researched theoretical concepts. The Uses and Gratifications theory has been rejuvenated in current research circles due to the Internet. This st udy shows that problem-solving styles, coupled with an indivi duals usage of the In ternet, do have an effect on variables like information recall. This study is thus opening doors to new exploration and findings in communications and agricultu ral communications research. This study also demonstrates the need for continued monitoring of information technologies and presentation of Extension in formation. While researchers continue to debate whether interactivity a ffects attitude and r ecall of information, these findings show no individual effects of inte ractivity on attitude and information recall when presenting information-driven content to a young adu lt population. While inte ractivity is being included on many emerging sites in Extensi on (e-Xtension, 2005), there are populations, such as innovators, who may retain inform ation better from the non-interactive earlier versions of online Extension information. It is also noted that for low users of the Internet, the novelty of interactivity attr acts and keeps the interest of users to increase their retention of information, as supporte d by the literature. Th ese findings encourage designers of information-driven sites to take inventory of how they are presenting their

PAGE 129

115 information to specific audiences. While these findings imply many new challenges in the battle to get Extension information out successfully to a variety of audiences, they also shed light on how communicators can better provide information that will be retained. While it may be assumed that more innovative people will want to see high levels of interactivity online, this study has shown that that may actually hinder their retention of information presented. The more adaptiv e individuals will actua lly do better with less structure and ambiguity when working online. Extension professionals must be cognizant of how they are addressing these problems on a site. Young adult audiences must have a good first experience with Extension information in order to ensure they will become lifetime consumers of the information and services provided. Based on these results, there is a defini te need to communicate Extension and agricultural messages online in a variety of formats. The Web is not a one-size-fit all type of environment. Individual needs and ps ychological make-up must be brought into consideration when developing content and the methods with which it is presented.

PAGE 130

116 APPENDIX A INSTRUMENTS INFORMED CONSENT Protocol Title: The Effect of Cognitive Probl em Solving Style and Level of Interactivity on Attitudes toward and Recall of Web-based Information Please read this consent document carefully befo re you decide to partic ipate in this study. My name is Emily Rhoades; I am a doctoral student in the Department of Agricultural Education and Communication at the University of Flor ida. Thank you for taking the time to participate in this study. Your particip ation is completely voluntary and will help to evaluate the effectiveness of online ma terials based on your personal problem-solving style. There is no penalty for not participati ng. If you choose to participate, you will answer items on a confidential survey that will take about 10 minutes to complete and will be emailed a link to a Website to look at. You can stop any time without penalty and you do not have to answer any question you do not wish to answer. All answers are confidential to the extent provided by law. There are no known risks associated with this study and there is no co mpensation, other than extra credit, or other direct benefit to you for participation. By turning in the survey you agree that you have read this statement and are aware of your rights. If you have any questions about this resear ch please contact the study supervisor, Dr. Tracy Irani or myself. The campus a ddress is 305 Rolfs Hall, PO Box 110540, Gainesville, and Fl 32611-0540. The phone num ber is (352) 392-0502. Questions about your concerns or rights can be directed to the UFIRB office, PO Box 112250, University of Florida, Gainesville and Fl 32611-2250. IRB #2006-U-0087

PAGE 131

117 Please take a few minutes to complete the following questions to the best of your ability. It is very important for the success of this study that all questions be completed. Responding will only take approximately 10 minutes, and your r esponses to the study will stay confidential. Your cooperation is greatly appreciated. 1. How many hours a day do you spend on the Internet (not including e-mail)? a. 1 or less d. 6-7 b. 2-3 e. 8 or more c. 4-5 2. How many different Websites do you vis it in an average session online? a. 1-2 d. 7-8 b. 3-4 e. 9 or more c. 5-6 3. Have you ever created a complete Website? a. Yes b. No 4. On a scale of 1-5, how would you rate your ability to design and post Websites? (1 being low to 5 being high) 1 2 3 4 5 None Low Moderate High 5. Please indicate how often you do the following each week : Download music 1 Never 2 3 Sometimes 4 5 Very Often Read a Blog 1 Never 2 3 Sometimes 4 5 Very Often Instant message 1 Never 2 3 Sometimes 4 5 Very Often Read Facebook or Myspace 1 Never 2 3 Sometimes 4 5 Very Often Watch videos 1 Never 2 3 Sometimes 4 5 Very Often Shop online 1 Never 2 3 Sometimes 4 5 Very Often Shop/sell on Ebay 1 Never 2 3 Sometimes 4 5 Very Often Use a search engine 1 Never 2 3 Sometimes 4 5 Very Often Work on WebCT or other online course 1 Never 2 3 Sometimes 4 5 Very Often

PAGE 132

118 6. When presented with multimedia on a Website (such as pictures, video, audio, Flash), I like to click on it to gain extra information. 1 2 3 4 5 Strongly Neutral Strongly Disagree Agree 7. Do you have your own __________? a. Blog Yes No b. Facebook page Yes No c. Myspace page Yes No d. Website Yes No e. other Yes No (please list: ) 8. Do you own a computer? a. Yes b. No 9. Please indicate the way in which you access the Internet most often: (circle only one each for school and home) When at school : When at home/apartment : a. Dial-up a. Dial-up b. High-speed access b. High-speed access c. Wireless access c. Wireless access d. Computer lab d. Computer lab 10. Please circle approximately how many times a week you use the following methods to get news/information: Newspaper 1 Never 2 3 Sometimes 4 5 Very Often Radio 1 Never 2 3 Sometimes 4 5 Very Often Television 1 Never 2 3 Sometimes 4 5 Very Often Internet 1 Never 2 3 Sometimes 4 5 Very Often Books 1 Never 2 3 Sometimes 4 5 Very Often Magazines 1 Never 2 3 Sometimes 4 5 Very Often 11. What is your preferred method to find information: (please only choose one) a. Newspaper b. Radio c. Television d. Internet e. Book f. Magazines

PAGE 133

119 Please circle one number on each line be low that best describes how you feel about Websites on the Internet. Numbers an d indicate strong feelings; and indicated weaker feelings; an d indicated you are undecided. 12. I feel that many Website s on the Internet are: Good 1 2 3 4 5 Bad Credible 1 2 3 4 5 Not Credible Unbiased 1 2 3 4 5 Biased Difficult to understand 1 2 3 4 5 Easy to understand Not Important 1 2 3 4 5 Important Not Interactive 1 2 3 4 5 Interactive Easy to find 1 2 3 4 5 Hard to find Beneficial 1 2 3 4 5 Not Beneficial Believable 1 2 3 4 5 Unbelievable Trustworthy 1 2 3 4 5 Not Trustworthy Accurate 1 2 3 4 5 Inaccurate 13. Please rank your level of agreement with the following statements: I would rather surf the Internet than do something else. 1 Strongly Disagree 2 3 4 5 Strongly Agree My knowledge increases as my Internet usage increases 1 Strongly Disagree 2 3 4 5 Strongly Agree It would be very difficult for me to survive without the Internet for several days. 1 Strongly Disagree 2 3 4 5 Strongly Agree Internet users are better-educated people. 1 Strongly Disagree 2 3 4 5 Strongly Agree The Internet opens doors that would otherwise be closed. 1 Strongly Disagree 2 3 4 5 Strongly Agree Information online should be engaging. 1 Strongly Disagree 2 3 4 5 Strongly Agree Information online should be interactive. 1 Strongly Disagree 2 3 4 5 Strongly Agree Information online should be entertaining. 1 Strongly Disagree 2 3 4 5 Strongly Agree 14. Gender

PAGE 134

120 a. Male b. Female 15. Age a. 18-20 b. 21-23 c. 24-27 d. 28 or older 16. Major _____________________________ 17. College___________________________ 18. Rank in school a. Freshmen b. Sophomore c. Junior d. Senior e. Graduate student This study is a two part study. The second par t of this study will be emailed to your gatorlink email if you wish to continue. Please indicate your gato rlink username to continue with this study: ____________________________________

PAGE 135

121 After you have looked through the site, please take a few minutes to complete the following questions to the best of your ability. Please do not return to the previous site. It is very important for the success of this study that all questions be completed. Responding will only take approximately 10 minutes, and your responses to the study will stay confidential. Your cooperation is greatly appreciated. Section 1: Introduction 1. Approximately how much time did you spend looking at the Website? 1-3 minutes 46minutes 7-9 minutes 10-12 minutes 13 or more minutes 2. Please rate your p revious knowledge of the topic presented on the Website. 1 Not Knowledgeable 2 3 Somewhat Knowledgeable 4 5 Very Knowledgeable How important was the information for you p ersonally? 1 Not Important 2 3 Somewhat Important 4 5 Very Important How motivated were you to read the information? 1 Not Motivated 2 3 Somewhat Motivated 4 5 Very Motivated How interested were you in the information p resented? 1 Very Interested 2 3 Somewhat Interested 4 5 Not Interested 3. 4. Indicate the number below that best descri bes how you feel about the term below. Boxes closer to the word indicate stro ng feelings and the middle box indicates that you are undecided. The information presented on the previously viewed Website is

PAGE 136

122 Good Bad Not credible Credible Biased Unbiased Difficult to understand Easy to understand Important Not Important Not interactive Interactive Easy to find Hard to find Not beneficial Beneficial Believable Unbelievable Not trustworthy Trustworthy Accurate Inaccurate 5. 6. Please indicate your level of agreement with the following: Content on this Website was interactive. 1 Strongly Disagree 2 3 4 5 Strongly Agree The content on this site kept me engaged. 1 Strongly Disagree 2 3 4 5 Strongly Agree I have used information p resented by the University of Florida Extension service. 1 Strongly Disagree 2 3 4 5 Strongly Agree I have looked at the University of Florida Extension Website. 1 Strongly Disagree 2 3 4 5 Strongly Agree 7. 8. I define interactivity online as:

PAGE 137

123 9. The site I saw included: Yes No a. Animated images b. Text c. Images d. Pop-up window Section 2: Car Buying e. 10. If given the opportunity to pur chase a car in the near fu ture, the likelihood of me doing so would be: Unlikely Likely Probable Improbable Definitely would not Definitely would 11. 12. Have you: Yes No a. Recently purchased a car b. Thought of purchasing a car c. Researched purchasing a car 13. Please list all facts you remember from the information you read previously: 14. Please describe what features you saw on this Website:

PAGE 138

124 15. Based on what you read on the previous Website: Which of these basic decisions doesn t need to be made when buying a car? Size of vehicle Safety features needed Fuel efficiency needed Color wanted 16. Based on what you read on the previous Website: Which of these is not a way to cut car insurance costs? Increasing the deductible amount for collision coverage Purchasing a newer car Paying annually or semiannually Dropping collision coverage 17. Based on what you read on the previous Website: When making the transportation decisions, whic h of these questions do you not need to ask? Should you trade in your car? Would you save money by taking a bus? Do you really need a car? How much gasoline would you need to buy? 18. Did you experience any problems while looking at the site? Yes -please explain No 19. Which course are you currently enrolled in? AEE 3030 Public Speaking AEE 3033 Technical Writing 20. Please list your gatorlink username to ensure your extra credit points: *Required Thank you for your time! S ubmit

PAGE 139

125 APPENDIX B EXPERIMENTAL CONDITION Non-Interactive Website Condition

PAGE 140

126

PAGE 141

127

PAGE 142

128 Interactive Website Condition

PAGE 143

129

PAGE 144

130

PAGE 145

131 LIST OF REFERENCES Abraham, L. (2001, August). Visual communication and new media: Defining visual communication in the era of convergence Paper presented at the annual conference of the Association for Educators in Journalism and Mass Communication, Washington, D.C. Agarwal, R. & Prasad, J. (1997). The role of innovation characte ristics and perceived voluntariness in the acceptance of information technologies. Decision Sciences, 28(3), 557-582. Agresti, A., & Finlay, B. (1997). Statistical methods for the social sciences. (Third Edition). Upper Saddle River, NJ: Prentice Hall. Amant, K. (2005). A prototype theory approa ch to International Website analysis and design. Technical Communication Quarterly, 14 (1), 73-91. Amponash, W.A. (1995) Computer adoption a nd use of information services by North Carolina commercial farmers. Jo urnal of Agricultural and Applied Economics, 27 (2), 565-576. Associated Press (2006). First-time car buyers seek out the Internet Retrieved May 12, 2006, from http://www.freep.com/apps/pbcs.d1 1/article?AID=/20060201/BUSINESS01/60201 0312/10 Ary, D., Jacobs, L.C., Razavieh, A. (2002). Introduction to research in education. Fort Worth, TX: Harcourt Brace College Publishers. Baozzi, R.P., & Foxall, G.R. (1995). Construc t validity and generalizability of the Kirton adaptation-innovation inventory. European Journal of Personality, 9 185-206. Baran, S.J., & Davis, D.K. (2003). Mass communication theory foundations, ferment, and future (3rd ed.). Belmont, CA: Wads worth/Thomson Learning. Bechtel, A. & Wu, H.D. (2002). Web site use and news topic and type. Journalism and Mass Communication Quarterly, 79(1), 73-86. Bezjian-Avery, A., Calder, B., & Iacobucci. (1998). New media inte ractive advertising vs. traditional advertising. Journal of Advertising Research, 38 (4), 23-32.

PAGE 146

132 Berger, S. (2001). Breaking up newsAn invest ment in the future? Correlations among hypertext comfort, user satis faction and perceived credib ility. Paper presented at the AEJMC Conference Papers virtual c onference. Retrieved June 27, 2005, from http://list.msu.edu/cgi-bin/wa? Berry, L.T. (2001). Comprehension and recall of Internet news: A quantitative study of Web page design. Journal of Magazine & New Media Research, 3 (1). Retrieved March 6, 2005 from http://aejmcmagazine.bsu.edu/jour nal/archive/Fall_2000/Berry3-1.html Bhate, S. (1999). Cognitive style differences and their impact on responses to message sources. Marketing Intelligence and Planning, 17 (6), 280-291. Blumler, J.G. (1979). The role of theory in Uses and Gratifications studies. Communication Research, 6 (1), 9-36. Bogart, L. (2000). The death of print. Editor & Publisher, 133(48). Boone, K., Meisenbach, T., & Tucker, M. (2000). Agricultural communications changes and challenges (First Edition), Iowa State University Press. Bouwman, H., &Van De Wijngaer t, L. (2002). Content and c ontext: An exploration of the basic characteristics of information needs. New Media and Society, 4 (3), 329353. Bressers, B. & Bergen, L. (2000, August). Internet use and media pr eferences of college students. Paper presented at the meeting of Ne wspaper Division of the Association for Education in Journalism and Mass Communication, Phoenix, AZ. Bull, N.H., Cote, L.S., Warner, P.D., & McKinn ie, M.R. (2004). Is Extension relevant for the 21st century? Journal of Extension, 42 (6). Retrieved September 29, 2005, from http://www.joe.org/joe/2004december/comm2.shtml Burns, E. (2005). The online battle of the sexes Retrieved May 20, 2006, from http://www.clickz.com/stats/sec tors/demographics/print.php/3574176 Camacho, M., Weinstock, D., & OGorman, K. (1997, August). A multi-method aesthetic approach to user-derived Internet interface designs. Paper presented at the meeting of Communication Technology and Policy Division of the Association for Education in Journalism and Ma ss Communication. Chicago, Ill. Campbell, D.T. & Stanley, J.C. (1963). Experimental and quasi-exp erimental designs for research. Chicago: Rand McNally. Cartmell, D.D., Dyer, J.E., Birkenholz, R.J., & Sitton, S. P. (2004). Publishing agricultural news: A st udy of Arkansas daily newspaper editors. Journal of Applied Communications, 87 (4), 7-22.

PAGE 147

133 Clemens, L. (2005). Courting youth Retrieved April 12, 2006, from http://marketingymedios.com/marketingym edios/magazine/article_display.jsp?vnu c Clement, J., Holbrook, P., & Staman, M. (1996). Extending Internet access to rural areas and small communities in the upper Midwestern United States. Retrieved December 20, 2002 from http://www.isoc.org/inet 96/proceedings/e5/e5_4.htm ClickZ. (2005). Three-quarters of Americans have access from home. Retrieved April 18, 2005, from http://www.clickz.com/news/article.php/3328091 Chan, J.K., & Leung, L. (2005). Lifestyles, relia nce on traditional news media and online news adoption. New Media & Society, 7 (3), 357-382. Chan-Olmsted, S.M., & Park, J.S. (2000). Fr om on-air to online world: Examining the content and structures of broa dcast TV stations Web sites. Journalism and Mass Communication Quarterly, 77 (2), 321-339. Chen, Q. (1999). Attitude toward the site. Journal of Advertising Research, 39 (5).Retrieved September 20, 2005, from EBSCO database. Chen, K., & Yen, D. C. (2004). Improving the quality of online presence through interactivity. Information Management, 42, 217-226. Cho, C. (1999). How advertising works on th e WWW: Modified el aboration likelihood model. Journal of Current Issues an d Research in Advertising, 21 (1), 33-50. Cho, C. (2003). The effectiveness of banne r advertisements: Involvement and clickthrough. Journalism and Mass Communication Quarterly 80 (3), 623-645. Christensen, L.B. (2001). Experimental methodology (8th ed). Boston, MA: Allyn and Bacon. Chung, H., & Zaho, X. (2004). Effects of perceived interac tivity on Web site preference and memory: Role of personal motivation. Retrieved May 18, 2006, from http://jcmc.indiana.edu/vol10/issue1/chung.html Collier, J.G. (2006). Cool gadgets luring young car buyers Retrieved April 12, 2006, from http://www.usatoday.com/money/autos /2006-03-28-caliber-smallcars_x.htm Conway, M. (2001). Cybernewswes, deserters, and includers. Pape r presented at the AEJMC Conference Papers virtual conference. Retr ieved June 27, 2005, from http://list.msu.edu/cgi-bin/ wa?A2=ind0109A&L=AEJMC&D=0&1=3&P=7755&F=P Conway, J. & Rubin, A. (1991). Psychologi cal predictors of television viewing motivation. Communication Research, 18, 443-463.

PAGE 148

134 Cyber Atlas (2003 ). February 2003 Internet Usage Stats Retrieved April 17, 2003, from http://cyberatlas. Internet.com/big_picture/tr affic_patterns/article/0,,5931_2169221, 00.htm. Danaher, P.J., & Mullarkey, G.W. (2003). Fact ors affecting online advertising recall: A study of students (Electronic Version). Journal of Advertising Research 43 (3). Davis, C. S., Akers, C., Cepica, M., Do erfert, D., Fraze, S., & Lawver, D. (2005, February). Cognitive responses by West Texas Hispanic/Latinos to agricultural news: A comparison of four Eng lish and Spanish presentation media. Paper presented at the annual meeting of the Southern Associati on of Agricultural Scientists, Little Rock, AK. DHaenens, L., Jankowski, N., & Heuvelman, A. (2004). News in online and print newspapers: Differences in reader consumption and recall. New Media & Society, 6 (3), 363-382. Diao, F., & Sundar, S. S. (2004). Orie nting response and memory for Web advertisements: Exploring effects of pop-up window and animation. Communication Research, 31 (5), 537-567. Dibean, W., & Garrison, B. (2001). How six online newspapers use Web technologies. Newspaper Research Journal, 22 (2), 79-93. Dillman, D.A. (2000). Mail and Internet surveys: The tailored design method (Second Edition), New York: John Wiley & Sons. Dimitrova, D.V, Connolly-Ahern, C., Williams A.P., Kaid, L.L., & Reid, A. (2003). Hyper linking as gate keeping: online ne wspaper coverage of the execution of an American terrorist. Journalism Studies, 4 (3), 401-414. Donaldson, J.L. (1998). What is Extension s itinerary for information superhighway travel. Journal of Extension, 36 (6). Downes, E.J., & McMillan, S. J. (2000) Defining interactivity: A qualitative identification of key dimensions. New Media & Society, 2 (2), 157-179. Dunn, C., Thomas, C., Green, C., & Mick, J. (2006). The impact of interactive multimedia on nutrition and physical activity knowledge of high school students. Journal of Extension 44 (2). Eagly, A.H., & Chaiken, S. (1993). The Psychology of Attitudes. Orlando, FL: Harcourt Brace Jovanovich, Inc. Eastin, M.S. (2001). Credibility assessments of online health information: The effects of source expertise and knowledge of content [Electronic version]. Journal of Computer-Mediated Communication 6 (4).

PAGE 149

135 Emery, M. (1999). Whos out there? St rengthening Internet communication for agriculture through consider ation of audience dimensions and user needs. Journal of Applied Communications, 83(1), 27-41. Esrock, S. & Leichty, G. (1999). Corporate world wide web pages: Serving the news media and other publics. Journalism and Mass Communication Quarterly, 76 (3), 456-467. Eveland, W.P., Cortese, J., Park, H., & D unwoody, S. (2004). How We b site organization influences free recall, factual knowle dge, and knowledge structure density. Human Communication Research, 30 (2), 208-233. Eveland, W.F., & Dunwoddy, S. (2001). User co ntrol and structural isomorphism or disorientation and cognitive load. Communication Research, 28 (1), 48-78. E-xtension. (2005). National extension imitative Retrieved October 8, 2005, from http://intranet.extension.org/index.php? module+articles&func=display&ptid=9&ai d=22 Fallows, D. (2004). The Internet and Daily Life Retrieved September 30, 2005, from http://www.pewInternet.org Fishbein, M. & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley. Fisher, S.G., Macrosson, W.D.K., & Wong, J. (1998). Cognitive style and team role preference. Journal of Manage rial Psychology, 13 (8), 544-556. Fox, J.R., Lang, A., Chung, Y., Lee, S., Schwar tz, N., & Potter, D. (2004). Picture this: Effects of graphics on the pro cessing of television news. Journal of Broadcasting & Electronic Media, 48 (4), 646-674. Foxall, G.R. (1996), Cognitive styles of consumer initiators. The Journal of Product Innovation Management, 13, (2), 172-173. Foxall, G. R. (1992). Gender differences in cognitive styles of MBA students in three countries. Psychological Reports, 70 169-170. Foxall, G., & Bhate, S. (1993). Cognitive styl e and personal involvement as explicators of innovative purchasing of healthy food brands. European Journal of Marketing, 27 (2), 5-16. Foxall, G., & Bhate, S. (1991). Psychology of computer use: XIX. Extent of computer use relationships with adaptive-i nnovative cognitive style and personal involvement in computing. Perceptual and Motor Skills, 72 195-202.

PAGE 150

136 Foxall, G., & Hackett, P.M. (1992). Cognitiv e style and extent of computer use in organizations: Relevance of sufficiency of originality, efficiency and ruleconformity. Perceptual and Motor Skills, 74 491-497. Foxall, G., & Haskins, C.G. (1986). Cognitiv e style and consumer innovativeness: An empirical test of Kirtons adaption-i nnovation theory in th e context of food purchasing. European Journal of Marketing, 20 (3/4), 63-80. Frick, M.J., Birkenholz, R.J., & Machtmes, K. (1995). Rural and urban adult knowledge and perceptions of agriculture. Journal of Agricultural Education, 36 (2), 44-53. Gallivan, M.J. (2003). The influence of soft ware developers cr eative style on their attitudes to and assimilation of a software process innovation. Information & Management, 40 (5), 443-465. Garrison, B. (2001). Diffusion of online in formation technologies in newspaper newsrooms. Journalism and New Technologies 2(2), 221-239. Goldsmith, R.E. (1984). Personality characteri stics associated with adaption-innovation. The Journal of Psychology, 117 159-165 Goldsmith, R.E., & Matherly, T.A.(1986). Seek ing simpler solutions: Assimilators and explorers, adaptors and innovators. The Journal of Psychology, 120 (2), 149-155. Graber, D. A. (1984). Processing the news: How peopl e tame the information tide. New York: Longman. Graziano, A.M., & Raulin, M.L. (2000). Research methods a process of inquiry. (Fourth Edition). Needham Heights, MA: Allyn & Bacon. Gruning, J.E., & Gruining, L.A. (1989). Toward a theory of public re lations behavior of organizations: Review of a program of res earch, in J.E. Gruining and L.A. Gruining (eds) Public Relations Research Annual vol. 1. Hillsdale, NJ : Erlbaum, 27-63. Gunter, B. (2000). Media research methods Thousand Oaks, CA: Sage Publications, 163-235. Gunter, B., Kinderlerer, J., & Beyleveld, D. (1999). The media and public understanding of biotechnology. Science Communication, 20 (4), 373-394. Hedges, L. E. (1991). Helping students de velop thinking skills through the problemsolving approach to teaching. The Ohio State University, Dr. Lowell Hedges. Heeter, C. (1989). Implications of interac tivity for communication research. In J.L. Salvaggio & J. Bryant (Eds.), Media use in the informati on age: Emerging patterns of adoption and consumer use (pp 217-235). Hillsdale, NJ: Lawrence Erlbaum.

PAGE 151

137 Henkia, A. (1990). Making the Web work for non-profits: Recommendations for the Ronald McDonald house of Dallas. Retrieved September 30, 2005, from http://lists.msu.edu/archives/aejmc.html Henroid, D., Ellis, J., & Huss, J. (2004). Me thods for answering food safety questions on the World Wide Web. Journal of Applied Communications, 87 (4), 23-34. Hoag, D.L., Ascough, J.C., & Frasier, W.M. (1999). Farm computer adoption in the Great Plains. Journal of Agricultural and Applied Economics, 31(1), 57-67. Hoffman, D.L., Novak, T.P., & Chatterjee, P. (2000). Commercial scenarios for the Web: opportunities and challenges [Electronic version]. Journal of Computer Mediated Communication, 1 (3 ). Horrigan, J.B., & Rainie, L. (2002). Counting on the Internet Retrieved September 30, 2005, from http://www.perwInternet.org Howell, J.L., & Harbon, G.B. (2004). Agricult ural landowners lack of preference for Internet Extension. Journal of Extension, 42 (6). Howell, J, Habron, G., & Woods, M. (2002, August). Water quality communications preferences of agricultural landowners. Paper presented at the Association for Agricultural Communicat ors in Education conference, Savannah, GA. Iowa State University (1998). Money mechanics: Owning a car Retrieved September 10, 2005, from http://www.extension.ias tate.edu/Publica tions/PM1461A.pdf Irani, T, & Kelleher, T. (1997, August). Information task equi vocality and media richness: Implications for health information on the World Wide Web. Paper presented at the Association for Educa tion in Journalism and Mass Communication conference, Chicago, Ill. Irani, T., Ruth, A., Telg, R. W., & Lundy, L.K. (2005, June). The ability to relate: Assessing the influence of a relationship ma rketing strategy and message stimuli on consumer perceptions of Extension. Paper presented at the Agricultural Communication Excellence conference, San Antonio, TX. Jackson, S.W., Hopper, G.M., & Clatterbuc k, W.K. (2004). Developing a national Webbased learning center for na tural resource education. Journal of Extension, 42 (1). Jensen, J.F. (1998). Interac tivity: Tracking a new concept in media and communication studies [Electronic version]. Nordicom Review, 19 (1), 185-204. Johnson, T.J., & Kaye, B.K. (2004). For who the Web toils: how Internet experience predicts Web reliance and credibility. Atlantic Journal of Communication, 12 (1), 19-45.

PAGE 152

138 Johnson, T.J., & Kaye, B.K. (1998). Cursing is believing?: Comparing Internet and traditional sources on media credibility measures. Journalism and Mass Communication Quarterly, 75 (2), 325-340. Katz, E., Blumler, J.G., & Gurevitch, M. (1974). Utilization of mass communication by the individual. In Blumler, J.G. (Eds.), The uses of mass communications. Beverly Hills, CA: Sage Publications. Kaslon, L., Lodl, K., & Greve, V. (2005). On line leader training for 4-H volunteers: A case study of action research. Journal of Extension, 43 (2). Retrieved September 29, 2005, from http://www.joe.org Kaye, B.K., & Johnson, T.J. (2002). Online and in the know: Uses and gratifications of the Web for political information. Journal of Broadcasti ng & Electronic Media, 46 (1), 54-71. Kelleher, T. (2001). Public relations tools and media choice. Journal of Public Relations, 13 (4). 303-320. Kelleher, T., Henley, M., Gennarelli, D., & Hon, L. (1997). Communication on the World Wide Web designing an effective homepage. Journal of Applied Communications, 81(2), 29-42. Kiousis, S. (2002). Interactiv ity: A concept explanation. New Media & Society, 4 (3), 355383. Kiousis, S. (2001). Public trust or mistrust ? Perceptions of media credibility in the information age. Mass Communication & Society, 4 (4). King, D.A., & Boehlje, M.D. (2000). Extens ion: On the brink of extinction or distinction? Journal of Extension, 38 (5). Retrieved May 5, 5005, from http://www.joe.org/joe/200october/comm1.html Kirton, M.J. (2003). Adaption-Innovation in the c ontext of diversity and change. New York, NY: Rutledge. Kirton, M.J. (1999). Kirton Adaption-Innovat ion inventory manual (3rd Ed.). New market, Suffolk UK: Occupational Research Centre. Kirton, M. J. (1989). Adaptors and Innovators: styles of creativity and pr oblem solving. New York, NY: Rutledge. Ko, H. (2002). A structural equation model of the Uses and Gratifications theory: Ritualized and instrume ntal Internet usage. Message posted at hhtp://list.msu.edu/archives/aejmc.html Ko, H. (2001). Internet Uses and Gratifications: understanding motivations for using the Internet. Message posted to hhtp://lis t.msu.edu/archives/aejmc.html

PAGE 153

139 Ko, H., Cho, C., & Roberts, M. (2005) Internet Uses and Gratifications. Journal of Advertising, 34 (2), 57-70. Kotrlik, J.W., & Williams, H.A. (2003). The incorporation of effect size in information technology, learning and pe rformance research. Information Technology, Learning, and Performance Journal, 21 (1), 1-7. Krug, S. (2000). Dont make me think. Indianapolis, IN: Circle.com Library. Kwang, N.A., Ang, R.P., Ooi, L.B., Shin, W.S., Oei, T.P.S., & Leng, V. (2005). Do adaptors and innovators s ubscribe to opposing values? Creativity Research Journal, 17 (2 & 3). 273-281. Ladner, M.D., Wingenbach, G.J., & Raven, M.R. (2002, February) Test of a bimodal survey model on the cooperative communi cators association: A case study. Paper presented at the meeting of Southern Agri cultural Education Research Conference, Orlando, FL. Lang, A., Borse, J., Wise, K., & David, P. (2002). Captured by the World Wide Web: Orienting of structural and content feat ures of computer-presented information. Communication Research 29 (3), 215-245. Lieb, R. (2005). Most Americans have PCs and Web access. Retrieved May 20, 2006, from http://www.clickz.com/stats/s ectors/geographics/print.php/3559991 Lin, C.A, & Jeffers, L.W. (2001). Compari ng distinctions and similarities across Websites of newspapers, radio st ations, and television stations. Journalism and Mass Communication Quarterly 78(3), 555-573. Lippert, R.M., Plank, O., & Radhakrishna, R. (2000). Beyond perceptio n: A pretest and posttest evaluation of a regional Inte rnet Extension in-service training. Journal of Extension 38 (2). Retrieved September 29, 2005, from http://www.joe.org Liu, Y. (2003). Developing a scale to m easure the interactivity of Websites. Journal of Advertising Research, 43 (2). Retrieved May 18, 2006, from EBSCO database. Lowery, W. (2002). The non-linear Web stor y: Assessment of perceptions, knowledge acquisition and feedback. Message posted to http://www.aejmc.com Lundy, L., Ruth, A., Telg, R., & Irani, T. (2005, February). It takes two: Public understanding of agricultural science and agr icultural scientists understanding of the public. Paper presented at the meeting of S outhern Association of Agricultural Scientists Agricultural Communications Section, Little Rock, AK. MacKenzie, S.B., & Lutz, R.J. (1989). An empirical examination of the structural antecedents of attitude toward the ad in an advertising pre-testing context. Journal of Marketing, 53 48-65.

PAGE 154

140 Maddox, S.J. (2001). Determining effective communicati on strategies for agricultural organizations to provide agricultural producers the knowledge necessary to promote change in the 21st century. Doctoral Dissertation NC State University Raleigh, NC. Maddox, S.J., Mustian, R.D., & Jenkins, D.M. (2003, February). Agricultural information preferences of North Carolina farmers. Paper presented at the meeting of Southern Association of Agricultural Scientists Agricultural Communications Section, Mobile, AL. Massey, B.L. (2004). Examination of 38 Web newspapers shows nonlinear storytelling rare. Newspaper Research Journal, 25 (3), 96-102. McDowell, G.R. (2001). Land-Grant Universities and Extension into the 21st Century. Ames, Iowa: Iowa State University Press. McLeod, J.M., & Becker, L.B. (1974). Testing the validity of gra tification measures through political effects analysis In Blumler, J.G. (Eds.), The Uses of Mass Communications. Beverly Hills, CA: Sage Publications. McMillan,S.J., & Hawng, J. (2002). Measures of perceived interactiv ity: An exploration of the role of direction of communicati on, user control, and time in shaping perceptions of interactivity. Journal of Advertising, 31 (3), 29-43. Melgares, P. (2005). Using feedback panels to analyze a web site's target audiences. Journal of Applied Communications, 89 (4), 9-20. Miller, J.D., Annou, M., & Wailes, E.J. (2004). Communicating biotechnology: Relationships between tone, issues, and te rminology in U.S. print media coverage. Journal of Applied Communications, 87 (3), 29-40. Moore, J. (2004, August). Information proces sing differences be tween Internet and magazine advertisements moderated by sele ctive exposure. Pape r presented at the allual meeting of the Association fo r Educators in Journalism and Mass Communication, Toronto, Canada. Morris, M., & Ogan, C. (1996). The Internet as mass medium. Journal of Communication 46 (1), 39-50. MSState (2005). The e-Extension initiative. Retrieved October 8, 2005, from http://asred.msstate.edu/na tional/htm2/initiative.htm Nelkin, D. (1995). Selling science: How the pres s covers science and technology. New York, NY: W.H. Freeman. Ness, M., Gorton, M., & Kuznesof, S. (2002), The Student Food Shopper: Segmentation on the Basis of Attitudes to Stor e Features and Shopping Behavior. British Food Journal, 104 (7), 506-526.

PAGE 155

141 Nielsen, J. (2000). Designing Web usability. Indianapolis, IN: New Riders Publishing. OMalley, M., & Kelleher, T. (2002). Papaya s and pedagogy: Geogr aphically dispersed teams and Internet self-efficacy. Public Relations Review, 28 175-184. Padilla-Walker, L., & Poole, D.A. (2002). Me mory for previous recall: A comparison of free and cued recall. Applied Cognitive Psychology, 16 515-524. Park, T., & Mishra, A. (2003, July). Internet usage by farmers: Evidence from a national survey Paper presented at the meeting of the AAEA, Montreal, Canada. Papacharissi, Z., & Rubin, A.M. (2000) Predictors of Internet use. Journal of Broadcasting & Electronic Media, 44 (2). Retrieved June 20, 2005, from EBSCO database. Pawlick, T.F. (2001). The invisible farm. Chicago, Ill.: Burnham Inc., Publishers. Peled, T., & Katz, E. (1974). Media functions in wartime: The Israel home front in October 1973. In Blumler, J.G. (Eds.), The Uses of Mass Communications. Beverly Hills, CA: Sage Publications. Pershyn, G. (1994). Understanding natural creative process using the KAI Retrieved May 2, 2006, from http://www.kaicentre.com/creative.htm Petty, R.E., Cacioppo, J.T. (1979). Issue involve ment can increase or decrease persuasion by enhancing message-relevant cognitive responses. Journal of Personality and Social Psychology, 37 (190), 1915-1926. Rayburn, J. (1996). Uses and Gratifications: An inte grated approach to communication theory and research Mahwah, NJ: Lawerence Erlbaum Associates, Inc. Reisner, A., & Walter, G. (1994). Agricultural journalists assessments of print coverage of agricultural news. Rural Sociology, 59 (3), 525-537. Resnick, M.L, & Montania, R. (2003). Percep tions of customer service, information privacy, and product quality from semiotic de sign features in an online Web store. International Journal of Hu man-Computer Interaction, 16 (2), 211-234. Rice, E. R., & Atkin, C.K. (1989). Public communication campaigns (2nd Ed.). Newbury Park, CA: Sage Publications. Rodgers, S., & Chen, Q. (2002). Post-adoption attitudes to advertising on the Internet. Journal of Advertising Research, 42 (5). Retrieved October 19, 2005, from EBSCO database. Rosengren, K.E. (1974). Uses and Gratificati ons: A Paradigm Outlines, in J.G. Blumler and E. Katz (eds). The Uses of Mass Communications, Newbury Park, CA: Sage..

PAGE 156

142 Rubin, A.M. (1994). The uses-and-gratificati ons perspective of media effects. In Ruggiero, T.E. (2000). Uses and gratifications theory in the 21st century Mass Communication & Society, 3 (1), 3-37. Runett, R. (2000). Study: Joint newsrooms still dominate. Retrieved May, 14, 2003, from http://www.digitaledge.org/mont hly/2000_04/mediaincyberspace.html Ruth, A., Bortree, D., Ford, R., Br aun, S., & Flowers, K. (2004, June). A new direction for agricultural media rela tions: Meeting journalists information needs through the Web. Paper presented at the Nationa l Association for Communication Conference, Lake Tahoe, NV. Ruth, A., Telg, R., Irani, T., & Locke, D. (2004, February). Agricultural scientists perceptions of fairness and accuracy of science and agriculture coverage in the news media. Paper presented at the meeti ng of Southern Association of Agricultural Scientists Agricultural Co mmunications Section, Tulsa, Oklahoma. Saunders, C., Akers, C., Haygood, J., & Lawver, D. (2003) World Wide Web coverage of agricultural issues: A content analysis. Proceedings of the Southern Association of Agricultural Scientists, Agricu ltural Communications Section, February 2003, Mobile, Al. Seevers, B., Graham, D., Gamon, J., & Conklin, J. (1997). Education through cooperative Extension Albany, NY: Delmar Publishers. Shortland, M., Gregory, J. (1991). Communicating science: A handbook. New York: Wiley. Sicilia, M., Ruiz, S., & Munuera, J.L. (2005) Effects of interactivity in a Web site. Journal of Advertising, 34 (3), 31-45. Siegrist, H., Labarge, G., & Prochaska, S. (1998). Using electronic media to convey timely information. Journal of Extension, 36(5). Retrieved December 4, 2002, from http://www.joe.org/1998october/iw1.html Sinclair, J., & Irani, T. (2005). A dvocacy advertising for biotechnology. Journal of Advertising, 34 (3), 59-73. Skinner, N.F., & Drake, J.M. (2003). Behavi oral implications of Adaption-Innovation: III. Adaption-innovation, achievement motivation, and academic performance. Social Behavior and Personality, 31 (1), 101-106. Slater, M.D., & Rouner, D. (1996). How messa ge evaluation and source attributes may influence credibility assessment and belief change. Journalism and Mass Communication Quarterly, 73 (4), 974-991.

PAGE 157

143 Spool, J.M. (1999). Website usability: A designers guide. San Diego, CA: Academic Press. Stanyer, J. (2001). The new media and the old: The press, broadcasting and the Internet. Parliamentary Affairs 54, 349-359. Stempel, G.H., Hargrove, T., & Bernt, J.P. (2000). Relation of growth of use of the Internet to changes in media use from 1995 to 1999. Journalism and Mass Communication Quarterly 77(1), 71-79. Stone, G., Singletary, M., & Richmond, V. P. (1999). Clarifying communication theories. Ames, IA: Iowa State University Press. Strover, S. (2001). Rural Internet connectivity. Telecommunications Policy, 25. 331-347. Suvedi, M., Campo, S., & Lapinski, M.K. (1999). Trends in Michigan farmers information seeking behaviors and perspe ctives on the delivery of information. Journal of Applied Communications 83(3), 33-50. Sundar, S.S. (2000). Multimedia effects on processing and perception of online news: A study of picture, audio, and video downloads. Journalism and Mass Communication Quarterly 77 (3). 480-499. Sundar, S.S., Kalyanaraman, S., & Brown, J. (2003). Explicating Web site interactivity: Impression formation effects in political campaign sites. Communication Research, 30 (1), 30-59. Sundar, S. & Nass, C. (2001, March). C onceptualizing sources in online news. Journal of Communication, 52-72. Swanson, D.L. (1979). The continuing evolution of the Uses and Gratifications approach. Communication Research, 6 (1), 3-7. Sweet, S.A. & Grace-Martin, K. (2003). Data analysis with SPSS: A first course in applied statistics. Boston, MA: Allyn and Bacon. Swenson, J., Constantinides, H., & Gurak, L. (2002). Audience-driven Web design: An application to medical Web sites. Technical Communication, 49 (3), 340-352. Taylor, J. (1993). KAI used with teenagers Retrieved May 2, 2006, from http://www.kaicentre.com/teen2.htm Teft, M. (1994). KAI and teenagers Retrieved April 26, 2006, from http://www.kaicentre.com/teenagers.htm Telg, R., Basford, A., & Ira ni, T. (2005, February). Communication preferences of politically active agricultural leaders. Paper presented at the annual meeting of the Southern Association of Agricultur al Scientists, Little Rock, AK.

PAGE 158

144 Teo, H., Oh, L., Liu, C., & Wei, K. (2003). An empirical study of the effects of interactivity on Web user attitude. International Journal Human-Computer Studies, 58 281-305. Terry, R., Jr. & Lawver, D. E. (1995). University students' perceptions of issues related to agriculture. Journal of Agricultural Education, 36 (4), 64-71 Tewksbury, D., & Althaus, S.L. (2000). Differe nces in knowledge acquisition among readers of the paper and online versi ons of a national newspaper. Journalism and Mass Communication Quarterly, 77 (3), 457-479. Thompson, D.R., & Wassmuth, B. (1999, August). Do they need a trick to make us click? Paper presented at the annual meeting of th e Association of Edu cators in Journalism and Mass communications, New Orleans, La. Till, B.D., & Baack, D.W. (2005). Recall and persuasion. Journal of Advertising, 34 (3), 4757. Tollett, J., Williams, R., & Rohr, D. (2002). Web design workshop. Berkeley, CA: Peachpit Press. Treise, D., & Weigold, M.F. (2002) Advancing science communication. Science Communication, 23 (3), 310-322. Trevino, L.K., Daft, R.L., & Lengel, R.H. ( 1990). Understanding managers media choices: A symbolic interactions perspective. In Fulk, J., & Steinfield, C. (Eds.), Organizations and Communication Technology Newbury Park, CA: Sage Publications. Tremayne, M. (2004). The Web of context: Appl ying network theory to the use of hyperlinks in journalism on the Web. Journalism and Mass Communication Quarterly, 81 (2), 237253. U.S. Confirms Mad Cow Case (2005, June). Retrieved October 20, 2005, from http://Web.lexisnexis.com/universe/document?_m=ebc36c157bf67ae1a930be235d081f7 Warner, P.D., Christenson, J.A., Dillman, D.A., & Salant, P. (1996). Public perception of Extension [electronic version], Journal of Extension, 3 4(4). Webster, J.G., & Lin, S. (2002). The Intern et audience: Web use as mass behavior. Journal of Broadcasting & Electronic Media, 46 (1), 1-12. Wicks, R.H. (1995). Remembering the news: Ef fects of medium and message discrepancy on news recall over time. Journalism and Mass Communication Quarterly, 72 (3), 666681.

PAGE 159

145 Williams, R. & Woods, M. (2002). A synthesis of agricultural communication research published in the Journal of Applied Communications from 1992-2001. Proceedings of the National Agricultural Communicato rs in Education Conference, 26-42. Wood-Turley, S. & Tucker, M. (2002). Measur ing preference for an agricultural college newsletter: A readership assessment of Missouris Discover & Enlighten Proceedings of the National Agricultural Commu nicators in Education Conference, 172-186. Wu, G. (1999). Perceived interactivity and att itude toward Website. Paper presented at the annual conference of the American Acad emy of Advertising, Albuquerque, NM. Wu, T., Custer, R.L., & Dyrenfurth, M.J. (1996). Technological and personal problemsolving styles: Is there a difference? Journal of Technology Education, 7 (2), 55-71.

PAGE 160

146 BIOGRAPHICAL SKETCH The author was born Emily Brin Bisdor f on October 1, 1980, in Columbus, Ohio. She grew up in Centerburg, Ohio, a small rura l town in central Ohio, where she graduated from Centerburg High School in 1998. Her love for livestock and agriculture stemmed from many years exhibiting sheep through 4-H and FFA. Emilys college career began in August of 1998 at The Ohio State University while serving as the Ohio FFA State Treasurer. Wh ile pursing her Bachelors of Science degree in agricultural communications, she spent a summer studying abroad at the Prague College of Agriculture in the Czech Republic. Emily spent her time working as an editorial and exhibit design intern while pursuing on her undergraduate degree. She graduated in March 2002, as a top ten senior in the College of Food, Agriculture, and Environmental Sciences at The Ohio State University. After completing her bachelors degree, Emily married her husband Aaron Rhoades and moved to Gainesville, Florida, to pursu e her Masters of Science in agricultural communications. Upon completion of her masters, Emily entered her doctoral program in agricultural communication at the University of Florida with an emphasis in new media communications. During her degree program Emily taught courses in technical communication, Web and print design, and publ ic relations. She conducted research in new media, distance education, Web materi al evaluation, and critical thinking. She served as the student board member for the Agricultural Institute in 2005-2006.

PAGE 161

147 Emily is a member of the Agricultural Education and Communication Graduate Student Association, Agricultural Communi cators of Tomorrow, Alpha Tau Alpha, Gamma Sigma Delta Honorary, Associati on for Communication Excellence, North American College Teachers of Agriculture, and the Amer ican Horse Publications Council.