can interactivity make a difference

The Internet is increasingly being recognized for its po-tential for health communication and education. The per-ceived relative advantage of the Internet over othermedia is its cost-effectiveness and interactivity, which inturn contribute to its persuasive capabilities. Ironically,despite its potential, we are nowhere nearer understand-ing how interactivity affects processing of health infor-mation and its contribution in terms of health outcomes.An experiment was conducted to examine the effects ofWeb interactivity on comprehension of and attitudestowards two health Web sites, and whether individualdifferences might moderate such effects. Two sites onskin cancer were designed with different levels of inter-activity and randomly assigned to 441 undergraduatestudents (aged 18–26) at a large southeastern university.The findings suggest that interactivity can significantlyaffect comprehension as well as attitudes towardshealth Web sites. The article also discusses insights intothe role of interactivity on online health communica-tions, and presents implications for the effective designof online health content.

Introduction and Rationale

Current enthusiasm for the Internet as a health commu-nication tool is based on its growing popularity as a sourceof health information, and on assumptions about the capac-ity of interactive technologies to provide active environ-ments for health education and promotion. However,before the Internet’s potential can be fully realized, Riceand Katz (2001) write, it is important that “the insights ofsocial science research need to be brought to bear onthe new systems as they are configured, made available,implemented and used” (p. 2).

Research in ehealth has largely focused on five aspects:the quality of health information sources; health information

seeking behavior; the changing dynamic of patient-providercommunication; online social support groups; and onlineclinical and health interventions. Less attention has beenpaid to the Internet as a channel and what makes it poten-tially effective for communicating health information andfor improving health care and well-being.

Interactivity has been pinpointed as the key feature ofInternet technology that makes it a compelling communica-tion tool. Combined with its capacity to disseminate infor-mation to mass audiences, the Internet also has persuasivequalities traditionally attributed to interpersonal channels.This so-called “hybrid channel” (Cassell, Jackson, &Cheuvront, 1998) can provide messages individualized toparticular needs and interests of users, and can encourageactive processing of health information.

However, there is still a dearth of theoretically-drivenempirical studies providing support for or against thisassumption. Reviews of interactive health systems havefound that these were at times “superior to and at times nobetter than other media with respect to educational andhealth outcomes” (Street & Rimal, 1997, p. 9). Aside frommethodological issues, a general lack of consensus in howinteractivity is conceptualized has been cited as a reason forthese mixed findings. Other scholars have proposed that thecommunicative efficacy of interactive technologies might beinfluenced more by their match with comprehensionprocesses than with the dynamism of the media itself(Aldrich, Rogers, & Scaife, 1998; Narayanan & Hegarty,2002). Newhagen and Rafaeli (1996) suggest that taking acloser look at how individual users access interactive infor-mation systems (e.g., their cognitive skills, ability to solveproblems and form searches, etc.) will have a significantbearing on our ability to fully exploit the Internet’s potentialas a communication and persuasive medium.

This study aims to answer this basic question: What caninteractivity contribute to desired outcomes such as compre-hension or attitudes? Also, do individual differences moder-ate the potential effect of interactivity on these desiredoutcomes?

JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE AND TECHNOLOGY—58(6):766–776, 2007

Can Interactivity Make a Difference? Effects ofInteractivity on the Comprehension of and AttitudesToward Online Health Content

Mia Liza A. LustriaCollege of Information, Florida State University, 270 Louis Shores Building, Tallahassee, FL 32306-2100. E-mail: [email protected]

Received January 19, 2006; revised May 31, 2006; accepted May 31, 2006

© 2007 Wiley Periodicals, Inc. • Published online 13 February 2007 inWiley InterScience (www.interscience.wiley.com). DOI: 10.1002/asi.20557

JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE AND TECHNOLOGY—April 2007 767DOI: 10.1002/asi

In the following section, we discuss the increasing popu-larity of the Internet as a source of health information andthen explore theories that might explain its utility for healthcommunication.

The Rise of the Ehealth “Prosumer”

There is consistent evidence that the Internet has becomea major influence on individuals’ health and lifestyles. About95 million Americans (80% of adult Internet users) searchedfor a major health topic online in 2004, making health infor-mation seeking one of the most popular activities on theInternet next to e-mail and researching products or services(Fox, 2005). In an earlier study, Fox and Fallows (2003)indicated that the major reasons people went online forhealth-related reasons were to seek general health informa-tion, to help prepare themselves for appointments and majorprocedures, to share information, and to provide support.Recent reports have revealed that seekers have become morepurposive and active in their searches, and more interested inwellness information (e.g., diet, fitness, and exercise), medicaltreatments, and information about health insurance andspecific doctors and hospitals (Fox, 2005).

Moreover, Horrigan and Rainie (2002) reported that 80%of Internet users have high expectations about the Internet asa health information source. Online sources have also beenshown to have a direct impact on patients’ health-relateddecision making and on the way people interact with theircaregivers (Baker, Wagner, Singer, & Bundorf, 2003; Cline,2003; Cline & Haynes, 2001; Fox & Fallows, 2003; Krane,2005). As such, the Internet is an “emerging and potentiallypowerful influence on health” (Evers, Prochaska, Prochaska,Driskell, Cummins, & Velicer, 2003, p. 4).

So, what makes the Internet a potentially effective toolfor health communication and education?

The Concept of Interactivity

Interactivity has been identified as one of the definingcharacteristics that sets new media apart from traditionalmedia. Interestingly, conceptualizations of this term havebeen hotly contested, making it difficult to examine it empir-ically and to draw solid conclusions about the role it plays inpromoting a variety of desired outcomes (e.g., informationseeking, learning, persuasion, etc.).

Interactivity has been approached from four differentstances: the nature of the communication exchange (Burgoon,Bonito, Bengtsson, Cederberg, Lundeberg, & Allspach, 2000;Jensen, 1998; Rafaeli & Sudweeks, 1997), system or channelfeatures (Andrisani et al., 2001; Bezjian-Avery, Calder, &Iacobucci, 1998; Chou, 2003; Coyle & Thorson, 2001;Downes & McMillan, 2000; Ha & James, 1998; Massey &Levy, 1999), user’s perceptions and/or actions (Light &Wakeman, 2001; McMillan, 2000; McMillan & Hwang,2002; Newhagen & Cordes, 1995; Sundar, Kalyanaraman, &Brown, 2003; Tremayne & Dunwoody, 2001), and some com-binationof theabove (Heeter,1989,2000;Kiousis,2002;Liu&Shrum, 2002; McMillan, 1999; McMillan & Huang, 2002).

Most definitions agree that interactivity of new communi-cation technology can be defined along three main dimen-sions: reciprocity/communication exchange, active usercontrol, and synchronicity (Heeter, 1989, 2000; Liu &Shrum, 2002). Reciprocity or communication exchangerefers to the ability of media to allow two-way interactionthrough feedback input devices, and means to communicatewith the system, other users, or with the content providers.Active user control refers to the ability of the media to allowthe user active control over their online experience. For ex-ample, navigational tools such as hyperlinks allow users tocontrol the direction and amount of their information expo-sure. Interactive systems also allow users to self-pace theirlearning experience. Synchronicity refers to the amount oftime it takes for the system to allow feedback. This conceptis the most ambiguous of all three because, for example,while fast download times may positively affect users’ per-ceptions of online experiences, the asynchronicity of tech-nology such as e-mail (allowing delayed feedback) can alsobe seen as a relative advantage.

The following section explores how interactivity can berelated to theories of message processing, comprehensionand learning, and individual differences.

Interactivity and Message Processing

The lure of interactive media is attributed to its ability toengage audiences much more than traditional media are ableto. Information processing theories posit that persuasion oc-curs more successfully as a result of the internalization ofmessages rather than from simple information retention.The elaboration likelihood model of persuasion (Petty &Cacioppo, 1979, 1986; Petty, Cacioppo, & Goldman, 1981),hypothesizes that messages that are not only attended to butthat are elaborated upon, are more likely to produce greaterand more permanent attitude change, compared to messagesthat are attended to with less scrutiny.

Cassell et al. (1998) write that the “transactional andresponse-dependent nature” (p. 75) of Internet communica-tions underlines its persuasive capabilities. First of all,health information seeking is commonly goal-driven andthere is an underlying motivation to seek content to fulfillcertain information needs. Second, online information seek-ing requires cognitive effort on the part of the individual—itis receiver-driven and requires active participation. Usershave the ability to engage content willfully, purposivelychoosing links and structuring their learning experience tomatch both their needs and capabilities. It is transactionalbecause it provides the mechanism for users to get immedi-ate feedback synchronously or asynchronously either fromthe system itself, from the health information provider, orfrom other users (Cassel et al., 1998).

Interactivity and Learning

Learning theories in the tradition of constructivism(Bruner, 1966; Piaget, 1970; Salomon, 1979) also lendthemselves to the argument that active involvement,

768 JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE AND TECHNOLOGY—April 2007DOI: 10.1002/asi

information processing, and learner control are key elementsto learning or knowledge acquisition.

Research in this area has shown how hypertext systemsmimic the associative nature of human information process-ing (Calisir & Gurel, 2003; Castelli, Colazzo, & Molinari,1998; Chen & Rada, 1996; Cho, 1995). Hypermedia schol-ars argue that users of nonlinear hypertext systems are ableto freely browse through a knowledge base, and redefineboth the structure and content of the material to be learned(Martindale, 1991; Nelson & Palumbo, 1992).

The cognitive flexibility theory (Spiro & Jheng, 1990)provides a rationale for the use of interactive technology, es-pecially for learning complex information. Spiro and Jheng(1990) define cognitive complexity as the “ability to sponta-neously restructure one’s knowledge . . . in adaptive re-sponse to radically changing situational demands . . . this isa function of both the way knowledge is represented and theprocesses that operate on those mental operations” (p. 165).Hypertext lends itself well to this type of learning becauseit allows learners to proceed freely through the system,randomly accessing material and processing informationaccording to individual mental models.

Not all hypertext systems have been found to enhancelearning, though. Poorly designed systems can either haveno effect on learning or lead to disorientation and cognitiveoverload (Baylor, 2001; Calisir & Gurel, 2003; Dias,Gomes, & Correia, 1999; McDonald & Stevenson, 1996,1998; Waniek, Brunstein, Naumann, & Krems, 2003; Zhang,Han, Zhu, & Zhu, 2002). These potential problems are partlyaddressed by providing good navigational aids (Chou, Lin,& Sun, 2000; Dias et al., 1999; Lee, 2002; Lee & Tedder,2004), although research has also shown individual differ-ences in how users react to variations in site organization.

The preceding discussion implies that the use of interactivetechnology may enhance learning and persuasion, but only tothe extent that the design of these systems carefully addressesissues that have been found to be important in traditionallearning environments. One common thread seems to linkthese various theoretical perspectives: that cognition andelaboration, plus active user involvement and control, areimportant precursors of desired outcomes. Thus, we predict

Hypothesis 1: Higher levels of interactivity will lead togreater comprehension of the content of a complex healthWeb site.

Hypothesis 2: Higher levels of interactivity will lead to morepositive attitudes towards the health Web site.

Interactivity, Comprehension, and Individual Differences

Despite the hype, there is still a dearth of theoreticallydriven empirical studies providing evidence for or againstthe notion that interactivity contributes to learning, compre-hension, and/or persuasion (Aldrich, Rogers, & Scaife,1998; Bezjian-Avery et al., 1998; Burgoon et al., 2000;Cairncross & Mannion, 2001; Chou, 2003; Dillon & Gabbard,1998; Evans & Sabry, 2003; Evers et al., 2003; Fiore &

Jin, 2003; Jaffe, 1997; Jimison, Adler, Coye, Mulley, & Eng,1999; Liu & Shrum, 2002; McMillan, 1999; Narayanan &Hegarty, 2002; Pavlou & Stewart, 2000; Reeves & Nass,2000; Schacter & Fagnano, 1999; Stout, Villegas, & Kim,2001; Street & Rimal, 1997). These mixed findings havebeen attributed to a general lack of consensus in how inter-activity has been operationalized (Ha & James, 1998;Heeter, 2000; Kiousis, 2002; McMillan & Hwang,2002; Rafaeli & Sudweeks, 1997; Steuer, 1992; Sundar,Kalyanaraman, & Brown, 2003) and to methodologicalissues (e.g., small sample sizes, confounding variables, etc.).

Narayanan and Hegarty (2002) propose an alternative viewto explain these mixed findings: that the “communicative eff-icacy of multimodal presentations may be more related to theirmatchwithcomprehensionprocesses thanwith the interactivityand dynamism of the presentation media itself” (p. 279).

In separate but similar studies, Lawless and Kulikowich(1994), Dillon (1991), and Caliser and Gurel (2003) foundthat individuals with higher previous knowledge demon-strated greater comprehension and were able to navigatenonlinear hypertext systems with less difficulty compared tothose with less previous knowledge.

A number of studies have suggested that recall of infor-mation was greater in traditional print sources compared toonline sources (Eveland Jr. & Dunwoody, 2002; Eveland,Cortese, Park, & Dunwoody, 2004; Sundar, Narayanan,Obregon, & Uppal, 1998; Tewksbury & Althaus, 2000;Tremayne & Dunwoody, 2001). These studies showed thatmore complex nonlinear hypertext structures seemed to en-courage more selective scanning of the text. Thus in nonlin-ear hypertext environments, readers were more likely to skipimportant information, compared to readers of print sources,who were generally disposed to read material from begin-ning to end. These studies also found that these differencesare mediated by information processing variables.

The preceding studies all point to an important juncture inonline communications research: the need to examine notonly the nature of online communications and its technicalaffordances, but also the need to examine how informationseekers process online content differently. The succeedingsection explores one individual difference that may influ-ence how people seek information online.

Need for Cognition as an Individual-Difference Variable

Need for cognition is one of the most studied individualfactors governing message processing, and indirectly, per-suasion (Petty & Cacioppo, 1986). High need for cognitionindividuals are those who enjoy thinking or who thought-fully consider information even when situational influencesdo not prompt such consideration (Bagozzi, Guerhan-Canli,& Priester, 2002). Initial studies of this construct have foundthat high levels of need for cognition are positively associ-ated with the tendency to scrutinize written communicationsmore carefully (Cacioppo, Petty, & Morris, 1983; Cohen,Stotland, & Wolfe, 1955); more active and greater informa-tion searches (Anderson, 2002; Chang & McDaniel, 1995;


Ketterer, 2001; Venkatraman, Marlino, Kardes, & Sklar,1990; Verplanken, Hazenberg, & Palenewen, 1992); a desireto engage in Web activities that require more effortful cogni-tive thought (Tuten & Bosnjak, 2001); deeper learning andhigher comprehension of complex course material (Diseth& Martinsen, 2003; Leone & Dalton, 1988); and betterdecision making strategies (Levin, Huneke, & Jasper, 2000;Smith & Levin, 1996), among others.

These studies imply a likely fit between this individual-difference variable and online health information seeking,which we have established earlier to be an inherently effortfulactivity. Thus we predict

Hypothesis 3: Higher levels of need for cognition will leadto greater comprehension of the content of a complexhealth Web site.

Hypothesis 4: Higher levels of interactivity will lead togreater comprehension of the content of a complex healthWeb site as a function of need for cognition.

The Current Study

To address the foregoing hypotheses, we designed a Webexperiment to correlate comprehension and attitudes withtwo independent variables: interactivity and need for cogni-tion, using a 2 � 2 factorial design. The purpose of this de-sign was to evaluate the potential effects of interactivity oncomprehension of and attitudes towards online health con-tent from an individual differences perspective. The keyresearch questions were as follows: Do comprehensionscores significantly differ according to levels of interactivity?Do comprehension scores significantly differ among treat-ment groups (with different levels of interactivity) as a func-tion of need for cognition? Do attitudes towards the sitesignificantly differ according to levels of interactivity?

Method

Research Design

An experimental 2 (high need for cognition vs. low needfor cognition) by 2 (high interactivity vs. low interactivity)factorial design was used to test the hypotheses.

Sampling and recruitment procedures. Test subjects wererecruited from a convenience sample of college studentsfrom a large southeastern university and randomly assignedto one of two treatment groups described in the experimen-tal design.

Study procedures. The Web experiment was conducted atselected computer laboratories on campus. Upon giving theirconsent to participate in the study, students were directed toa survey site, which detailed instructions for the rest of theactivity. Subjects were told to imagine that they had to findinformation about skin cancer for a friend or family memberwho had developed the disease. They were given approxi-mately 20 minutes to explore one of two test sites (randomly

assigned to subjects) and then another 30 minutes to answeran online survey.

Operational definitions and measures. All multiple-itemscales used 5-point Likert response formats. Items were re-coded as necessary so higher scale scores indicated higherendorsement of variables.

Demographics. Individuals were asked general demo-graphic questions including gender, race/ethnicity, and age.

Computer competency and Internet use. A 12-itemscale was used to measure respondents’ ability to do basiccomputer and Internet activities. Nine of the items on thescale were adapted from Swinyard and Smith’s (2003) com-puter competency scale (coefficient alpha of 0.90).

For this study, respondents were asked to evaluate howcompetent they were in doing various computer/Internet ac-tivities using a 5-point Likert-type scale, ranging from 1 (notcompetent at all), to 5 (very competent). The scale was veryreliable with a Chronbach alpha of 0.89.

Interactivity. Two aspects of interactivity were measured:perceived interactivity and technical interactivity. Technicalinteractivity referred to the formal features that were ma-nipulated in the stimulus materials. Two Web sites weredesigned using essentially the same text and graphicalcontent from credible online sources on skin cancer (withpermission from the original sites). The high-interactivitysite used a nonlinear hypertext structure and included vari-ous navigation tools, hyperlinks, and a few interactive ac-tivities (e.g., click-through modules, animation, interactivequizzes, and tailored queries). The low-interactivity siteused a hierarchical linear hypertext structure with minimalnavigation. The primary dimensions of technical interactiv-ity that were explored in this study were active user control,sensory stimulation, and synchronicity.

Another measure of interactivity that was of interest inthis study was perceived interactivity. This was measuredusing selected items from the active control and synchronic-ity subscales of Liu’s (2003) perceived interactivity scale.Respondents were asked to signify their agreement, usingratings from 1 (strongly disagree) to 5 (strongly agree) withfive items including statements such as “I felt I had a lot con-trol over where I wanted to go on the site,” and “When Iclicked on links I felt I was getting instantaneous informa-tion.” The scale was reliable with a coefficient alpha of 0.77.

Need for cognition. Need for cognition was measuredusing the 18-item need for cognition scale developed byCacioppo, Petty, and Kao (1984). Respondents were askedto indicate their agreement (with 5 meaning strongly agree)with statements such as “I would prefer complex to simpleproblems,” “I like to have the responsibility of handing a sit-uation that requires a lot of thinking,” and “thinking is notmy idea of fun.” The coefficient alpha for the scale was 0.87.


Comprehension. Comprehension referred to the correctencoding (or understanding) of the text and was defined asthe amount of meanings accurately drawn from the message.Comprehension was assessed using eight modified true-or-false items with three response options: true, false, and Idon’t know. The final 8-item comprehension scale was culledfrom an original pool of 23 items that were pretested to de-termine each item’s level of difficulty, as well as to ensurethat the items measured understanding of only material-asserted concepts. These items were later analyzed to ensurethat they were neither too easy nor too difficult to under-stand, and to ensure that they adequately discriminated be-tween high and low scorers.

Attitude towards the site. Attitude was conceptualized asan evaluative response to the stimulus material—the Web site.Participants indicated their agreement with such items as “Thesite was attractive,” “I enjoyed exploring the site,” and “Iwould visit the site again in the future.” Reliability of this scalewas indicated by a Chronbach’s coefficient alpha of 0.86.

Reading style. Respondents were asked to characterizehow carefully they read the site content. This was measuredat three levels: “I read the whole content,” “I skimmed thecontent and only read carefully items that interested me,” “Ijumped sections and only read items that caught my eye.”

Time on task. Time on task was defined as the amount oftime (in minutes) the respondents spent browsing the testsite assigned to them.

Results

Sociodemographic Characteristics of the Respondents

About 441 undergraduates comprised the test sample. Ofthese, 44% (n � 194) were male, and 53.3% (n � 235) werefemale. The sample was limited to a certain age group(young adults aged 18–26 years); the mean age was 19 yearsold. The majority of the respondents were white (84.1%),while a minority were African-American (5.4%) or of an-other race (4.8%). A great majority of the respondents hadaccess to the Internet from their homes (95.9%), have hadsix or more years of Internet experience (69.6%), and usedthe Internet most frequently from their homes (M � 4.58) ortheir schools (M � 4.17). The sample was moderately

computer literate with a mean computer competency scoreof 3.79 (Table 1).

Manipulation Checks

To test the effectiveness of the experimental manipula-tion, a one-way analysis of variance (ANOVA), with per-ceived interactivity as the dependent variable and level oftechnical interactivity as the independent variable, was per-formed. As expected, the high-interactivity group ratedtheir site as more interactive (M � 3.98, SD � 0.58) com-pared to the low-interactivity group (M � 3.35, SD � 0.78).Results of the ANOVA (as shown in Table 2) show thatthese differences were highly significant, F(1,439) � 94.98,p � 0.000, partial �2 � 0.18. This analysis shows that the

TABLE 1. Sociodemographic characteristics and computer or Internetuse of respondents (Overall).

Variable N � 441 % M SD

GenderMale 194 44.00Female 235 53.30

Age (range 18–26) 428 19.30 1.16

RaceWhite or Caucasian 374 84.10African-American 24 5.40Other/Multiracial 21 4.80

Have access to Internet from homeYes 423 95.90No 18 4.10

Computer Competency Scorea 441 3.79 0.71

Years Using the InternetNever used the Internet 2 0.50< 1 year 1 0.201 year 1 0.202 years 8 1.803 years 14 3.204 years 28 6.305 years 69 15.606 years or more 307 69.60

Frequency of Using 441Internet fromb. . .Home 4.58 0.96School 4.17 1.06Work 2.16 1.60

aHighest mean score was 5.bRated with 5 as the highest.

TABLE 2. One-way analysis of variance summary for perceived interactivity.

Source df SS MSE F p-value Partial �2 Observed powera

Between groups 1 44.65 44.65 94.98 0.000 0.18 1.00Within groups 439 206.40 0.47Total 441 6190.32

aComputed using alpha � 0.05.


high-interactivity site was indeed perceived to be more in-teractive than the low-interactivity site.

To check for potential usability issues related to the dif-ferent hypertext structures adapted in the two sites, a secondmanipulation check was conducted with perceived disorien-tation as the dependent variable, and interactivity as the in-dependent variable. Perceived disorientation was measuredby asking respondents to rate their agreement with the state-ment “At times I felt confused about where I was, where Iwas going, or where I had been” using a 5-point Likert-typescale, with 5 as the highest ranking.

The results show that the low-interactivity group ratedtheir site as slightly more confusing (M � 2.27, SD � 0.99)compared to the high-interactivity group (M � 2.21, SD �1.09). This difference (as shown in Table 3), however, was notsignificant, F(1,439) � 0.39, p � 0.53, partial �2 � 0.001.

Correlational Analysis

Pearson product moment correlations were computed be-tween the dependent variables (comprehension and attitudestowards the site) and six other variables: time on task, age,need for cognition, number of sources on skin cancer previ-ously read, health literacy, and computer competency. Re-sults show that 9 out of the 27 correlations were statisticallysignificant, although these signified overall weak relation-ships (Table 4). Time on task was found to have a weak butpositive relationship with comprehension (r (422) � 0.20,p � 0.01). Moreover, number of skin cancer sources previ-ously read was also found to have a weak but positive rela-tionship with comprehension (r (422) � 0.13, p � 0.01) andattitudes towards the site (r (422) � 0.21, p � 0.01).

On the other hand, need for cognition was found to have:a weak but positive relationship with time on task (r (422) �0.17, p � 0.01.), a weak but positive relationship with com-puter competency (r (422) � 0.15, p � 0.01), and a weakbut positive relationship with health literacy (r (422) � 0.13,p � 0.01).

Comprehension Scores, Treatment, and Covariates

Hypothesis 1 predicted that those assigned to the high-interactivity site would have higher comprehension scoresthan those exposed to the low-interactivity site. Resultsshow that respondents in the high-interactivity group hadhigher mean comprehension scores (M � 4.79, SD � 1.55for the high NFC group and M � 4.70, SD � 1.48 for thelow NFC group) than the low-interactivity group (Table 5).Results of the two-way ANCOVA show that the main effectof level of interactivity on comprehension scores was signif-icant, F(1,435) � 6.67, MSE � 14.92, p � 0.01; see Table 6.Based on these results, the hypothesis was supported.

TABLE 3. One-way analysis of variance summary for perceived disorientation.

Source df SS MS F p-value Partial �2 Observed powera

Between groups 1 0.43 0.43 0.39 0.53 0.001 0.10Within groups 439 4766.10 1.09Total 441 2690.00


TABLE 4. Correlational analysis among comprehension, attitudes, and six other variables of interest.

Time on Need for No. of sources on Health Computer Variables M SD task Age cognition skin cancer read literacy competency

Comprehension 4.67 1.55 0.20a �0.05 0.03 0.13a 0.05 �0.01Attitude Towards Site 3.57 0.75 �0.01 �0.06 0.08 0.21a �0.02 �0.16a

Time on Task 18.46 10.18 1.00 �0.12a 0.17a 0.03 0.10* �0.07Age 19.29 1.15 1.00 0.12a �0.07 �0.04 0.05Need for Cognition 3.18 0.54 1.00 0.05 0.13a 0.15a

No. of Sources on Skin 1.96 1.88 1.00 0.02 �0.09Cancer Read

Health Literacy 29.66 6.37 1.00 0.05Computer Competency 3.81 0.71 1.00

aCorrelation is significant at the 0.01 level (1-tailed); Listwise N � 422.

TABLE 5. Mean comprehension scores and standard deviations as a func-tion of level of interactivity and need for cognition.

High need for Low need for Source cognition cognition

High Interactivity M � 4.79 M � 4.70SD � 1.55 SD � 1.48

N � 103 N � 121Low Interactivity M � 4.44 M � 4.67

SD � 1.48 SD � 1.67N �103 N � 114


As assessed by the partial �2, however, the strength of therelationship between level of interactivity and comprehen-sion scores, holding constant time on task and reading style,was not very strong, with level of interactivity accounting foronly 1.5% of the variance in the mean comprehension scores.

Hypothesis 3 predicted that high need for cognition(NFC) individuals would have higher comprehension scoresthan those with low need for cognition. Table 5 shows mixedresults in that high NFC individuals exposed to high interac-tivity had the highest mean comprehension scores (M �4.79, SD � 1.55), but the high NFC individuals exposed tolow interactivity had the lowest mean comprehension scores(M � 4.44, SD � 1.48). Moreover, results of the two-wayANCOVA (Table 6) show that the main effect of need forcognition on comprehension scores controlling for time ontask and reading style was not significant, F(1,435) � 2.15,MSE � 4.82, p � 0.14, partial �2 � 0.005. Based on theseresults, Hypothesis 3 was not supported.

Hypothesis 4 predicted that those assigned to the high-interactivity site would have higher comprehension scoresthan those who were exposed to the low-interactivity site asa function of need for cognition. The mean comprehensionscores adjusted for by initial differences in time on task andreading style, however, was not as clearly ordered across thefour groups based on need for cognition and level of interac-tivity (as shown in Table 5). Results show, for example, thatthe low-interactivity, high NFC group (M � 4.44, SD � 1.48)had slightly lower mean comprehension scores than the low-interactivity, low NFC group (M � 4.67, SD � 1.67). More-over, results of the ANCOVA revealed that there was no sig-nificant interaction between level of interactivity and needfor cognition on mean comprehension scores holding con-stant time on task and reading style, F(1,435) � 1.80,

MSE � 4.03, p � 0.18, partial �2 � 0.004. Based on theseresults, Hypothesis 4 was not supported.

There is no clear indication why this occurred. We specu-late, though, that this may have been due to the fact thatthere just was not enough variability between groups onneed for cognition to show clearer and more significant dif-ferences between treatment groups on the dependent vari-able. We also suspect that the results would have beenclearer (and in the direction predicted) had the sample beenmore heterogeneous.

Attitude Towards Site and Treatment

Hypothesis 2 predicted that those assigned to the high-interactivity site would have more positive attitudes towardsthe site than those who were exposed to the low-interactivitysite. Results show that the mean comprehension scores werein the direction predicted (Table 7). Respondents in the high-interactivity group had higher mean comprehension scores(M � 3.71 vs. M � 3.45). Results of the two-way ANCOVA(Table 8) show that the main effect of level of interactivityon attitudes towards the site with reading style as a covariatewas highly significant, F(1,438) � 16.56, MSE � 8.86,p � 0.0, with a medium effect size of partial �2 � 0.036.

TABLE 6. Analysis of covariance of comprehension scores as a function of level of interactivity and need for cognition, with time on task and reading styleas covariates.


Covariate (Time on task) 1 28.66 28.66 12.82 .000 .029 .947Covariate (Reading style) 1 12.24 12.24 5.48 .020 .012 .646Level of Interactivity (I) 1 14.92 14.92 6.67 .010 .015 .732Need for Cognition (NFC) 1 4.82 4.82 2.15 .143 .005 .310I � NFC 1 4.03 4.03 1.80 .180 .004 .268Error 435Total 441


TABLE 7. Mean attitudes towards site and standard deviations as afunction of level of interactivity.

Treatment M SD N

Low interactivity 3.45 .749 217High interactivity 3.71 .724 224Total 3.58 .747 441

TABLE 8. Analysis of covariance of attitudes towards site as a function of level of interactivity with reading style as covariate.


Covariate (Reading style) 1 3.780 3.780 7.068 .008 .016 .756Level of Interactivity (I) 1 8.858 8.858 16.563 .000 .036 .982Error 438Total 441



Based on these results, Hypothesis 2 was supported. Theresults are discussed in the next section.

Discussion, Limitations, and Suggestions forFurther Study

Conceptually, Internet technologies hold much promisefor improving health knowledge and behavior. Despite theseexciting claims, we still know very little about how interac-tive Web technologies influence information use, learning,and motivational processes.

The main goal of this study was to examine the effects ofinteractivity on the comprehension of and attitudes towardsonline health content. Moreover, it sought to determinewhether individual differences in ability to elaborate (needfor cognition) would moderate the effects of interactivity.

The results of the main analysis show that the mean com-prehension scores were in the direction predicted: respon-dents exposed to high interactivity had higher mean compre-hension scores than those exposed to the low-interactivitysite. Furthermore, the results of the ANCOVA found thatthere was a significant main effect for level of interactivityon mean comprehension scores, controlling for time on taskand reading style. The results, however, did not show a sig-nificant main effect for need for cognition on comprehensionscores. Neither was there a significant interaction betweenneed for cognition and level of interactivity on comprehen-sion scores controlling for time on task or reading style.

At first glance, the significant main effect found for level ofinteractivity seems compelling. As hypothesized, respondentsin the high-interactivity group were able to comprehend theinformation better, theoretically because they had greater con-trol over their learning environment and because they had theopportunity to interact with more engaging learning activities(i.e., the interactive activities). However, this may be mislead-ing because the experiment had sufficient power to detectsuch a small effect size as evidenced by the partial �2 � 0.015for this particular main effect. Moreover, while the focus ofthe analysis was level of interactivity, we cannot discountthe significance of the two covariates controlled for in theANCOVA: time on task and reading style. Both had highersignificance values and partial �s than any of the independentvariables studied. So the statistical test provides support for acompelling main effect of level of interactivity, but this doesnot improve our understanding of why this is so.

On a positive note, though, the initial analysis of the data didreveal a significant relationship between level of interactivityand attitudes towards the site. This implies that interactivitymay play an important role in attracting health-informationseekers and in maintaining their attention. However, furtheranalysis is needed to reveal what particular features of a Website contribute to more positive attitudes towards the site.

Interactive systems may provide a whole range offeatures that can enhance learning of online content, but theyalso present new challenges to individuals who may be moreused to extracting meaning from traditional linear text. This

brings up one point of speculation: whether greater interac-tivity on a Web site might be a potential distraction for cer-tain types of users. In interactive environments, knowledgestructures are created on the fly by both the reader and the in-formation designer — the reader must not only identify whatinformation they need to enhance their comprehension of thematerial, but they must also know where to find this infor-mation. Researchers have found that these decisions addcognitive burden to information seekers who do not haveadequate domain knowledge or who are not interested in thecontent area (Alexander & Jetton, 2003; Calisir & Gurel,2003; Lawless & Brown, 1997; Lawless & Kulikowich,1994; Niederhauser, Reynolds, Salmen, & Skolmoski,2000). Moreover, online environments are infinitely mutableand adaptive; one user’s text (content) may be entirelydifferent from another user’s text. Individuals differ in theirstrategies to traverse the system, in their choices of whatelements to interact with, and in their interpretations of theoverall meanings of the information they encounter.

Several scholars suggest that there is a need to developspecific online competencies for navigating hypertext struc-tures and reading online content (Coiro, 2003; Detlefsen,2004; Kovacs, 2004; MacGregor, 1999; Unz & Hesse,1999). What exactly these competencies and skills shouldentail (i.e., learning styles, navigational styles, or cognitivestyles) is still poorly understood.

Implications for the Design of Health Information Web Sites

This study raises important issues about the accessibilityand usability of online health information sources. While thefindings provided evidence that interactivity can signifi-cantly increase comprehension of online health content, theliterature has also revealed that greater interactivity may infact create more difficulties for information seekers who areless competent in navigating online environments, who maybe less familiar with the content of the Web site, or who maybe less motivated to navigate the site in the first place. Iron-ically, while increased user control has been used to explainwhy high interactivity would theoretically lead to greatercomprehension, this was not supported fully in the contextof this study. The limitations of putting all control in thehands of information seekers include the following: theymay gloss over or totally miss reading relevant information,they may become confused and lost, or they may becomefrustrated with the site and quit browsing the site sooner.

What are the implications of this on the design of onlinehealth content? First of all, this brings to the fore an apparentdisconnect between the features information designers mayfind to be compelling about interactive media (e.g., high-tech features such as flash, animation, interactive activities)and what health seekers need (e.g., a clear path to simple an-swers to their questions). The challenge, therefore, is how tostrike a balance between providing an engaging and visuallyappealing Web site, and providing a site that the least com-petent information seeker could easily navigate without toomuch guesswork.


Certainly, interactivity brings to the table many newcapacities for health communication. On the other hand, itmay also provide many new challenges for various individu-als. Understanding differences in the way individuals processsimilar content delivered using different levels of interactiv-ity will also better inform us about how to design betterhealth information delivery systems and how to use interac-tivity more effectively for health education and promotion.

Limitations and Suggestions for Future Study

Conducting an experimental study of this type is notwithout its limitations. First of all, like most experimentalresearch, it would be difficult for us to claim how generaliz-able the findings might be. Interactivity here was narrowlydefined by features, and yet did not include all the possiblefeatures that could make Web sites truly interactive (e.g.,feedback devices, chat, BBSs, etc.). Perhaps the greatestgains from online health Web sites arise more from featuresthat support the socioemotional needs of health informationseekers rather than the technical features that make the In-ternet a dynamic communication channel. It would be inter-esting to study what draws different types of users to specifichealth Web sites, what features engage them the most, andwhich features are the most helpful in meeting their expecta-tions and needs.

Another limitation of the study was the homogeneity of thesample population. This study could be enriched by testingthe effects of level of interactivity on a more heterogeneoussample to introduce variability on a number of characteristics(e.g., age, race, socioeconomic status, educational attain-ment, computer competence, Internet experience, healthstatus, perceived risk for the disease, previous knowledge ofthe disease) that may directly or indirectly affect performancein online environments. This approach may lead to sugges-tions for how best to design online health information sys-tems for more disadvantaged and at-risk groups.

Acknowledgments

This research project was conducted at the University ofKentucky with funding from the Jacobs Foundation Informa-tion Technology Dissertation Grant (Zurich, Switzerland),the Beta Phi Mu Eugene Garfield Dissertation Fellow-ship (USA), and the University of Kentucky DissertationEnhancement Award (Lexington, KY, USA). Thanks areextended to Dr. J. David Johnson, Dr. Donald O. Case, andDr. Corinne Jorgensen for providing comments on earlierversions of this article.

References

Aldrich, F., Rogers, Y., & Scaife, M. (1998). Getting to grips with “interac-tivity”: Helping teachers assess the educational value of CD-ROMs.British Journal of Educational Technology, 29(4), 321.

Alexander, P.A., & Jetton, T.L. (2003). Learning from traditional and alter-native texts: New conceptualizations for the information age. In A.C.

Graesser & M.A. Gernsbacher (Eds.), Handbook of discourse processes(pp. 199–241). Mahwah, NJ, US: Lawrence Erlbaum Associates.

Anderson, M.H. (2002). The effects of individuals’ social network charac-teristics and information processing characteristics on their sensemakingof complex, ambiguous issues. Unpublished dissertation, University ofMinnesota, Twin Cities, Minnesota.

Andrisani, D., Gaal, A.V., Gillette, D., & Steward, S. (2001). Making themost of interactivity online. Technical Communication, 48(3), 309–323.

Bagozzi, R.P., Guerhan-Canli, Z., & Priester, J.R. (2002). The socialpsychology of consumer behaviour. Buckingham, England: OpenUniversity Press.

Baker, L., Wagner, T.H., Singer, S., & Bundorf, M.K. (2003). Use of the In-ternet and e-mail for health care information: Results from a nationalsurvey. Journal of the American Medical Association, 289(18),2400–2406.

Baylor, A.L. (2001). Perceived disorientation and incidental learning in aWeb-based environment: Internal and external factors. Journal of Educa-tional Multimedia and Hypermedia, 10(3), 227–251.

Bezjian-Avery, A., Calder, B., & Iacobucci, D. (1998). New media interac-tive advertising vs. traditional advertising. Journal of Advertising Re-search, 38(4), 23–33.

Bruner, J.S. (1966). Toward a theory of instruction. Cambridge, MA: Belk-nap Press of Harvard University.

Burgoon, J., Bonito, J.A., Bengtsson, B., Cederberg, C., Lundeberg, M., &Allspach, L. (2000). Interactivity in human-computer interaction:A study of credibility, understanding, and influence. Computers inHuman Behavior, 16(6), 553–574.

Cacioppo, J., Petty, R.E., & Kao, C.F. (1984). The efficient assessment ofneed for cognition. Journal of Personality Assessment, 48(3), 306–307.

Cacioppo, J. ., Petty, R.E., & Morris, K.J. (1983). Effects of need for cogni-tion on message evaluation, recall, and persuasion. Journal of Personalityand Social Psychology, 45(4), 805–818.

Cairncross, S., & Mannion, M. (2001). Interactive multimedia and learning:Realizing the benefits. Innovations in Education and Teaching Interna-tional, 38(2), 156–164.

Calisir, F., & Gurel, Z. (2003). Influence of text structure and prior knowl-edge of the learner on reading comprehension, browsing, and perceivedcontrol. Computers in Human Behavior, 19(2), 135–145.

Cassell, M.M., Jackson, C., & Cheuvront, B. (1998). Health communica-tion on the Internet: An effective channel for health behavior change?Journal of Health Communication, 3(1), 71–79.

Castelli, C., Colazzo, L.,& Molinari, A. (1998). Cognitive variables andpatterns of hypertext performances: Lessons learned for educationalhypermedia construction. Journal of Educational Multimedia and Hyper-media, 7(2–3), 177–206.

Chang, C.K., & McDaniel, E.D. (1995). Information search strategies inloosely structured settings. Journal of Educational Computing Research,12(1), 95–107.

Chen, C., & Rada, R. (1996). Interacting with hypertext: A meta-analysis ofexperimental studies. Human-Computer Interaction, 11(2), 125–156.

Cho, Y. (1995, June 17–19). Learner control, cognitive processes, and hy-pertext learning environments. Paper presented at the National Educa-tional Computing Conference, NECC ‘95, Baltimore, MD.

Chou, C. (2003). Interactivity and interactive functions in Web-basedlearning systems: A technical framework for designers. British Journalof Educational Technology, 34(3), 265–279.

Chou, C., Lin, H., & Sun, C.T. (2000). Navigation maps in hierarchical-structured hypertext courseware. International Journal of InstructionalMedia, 27(2), 165–182.

Cline, R.J.W. (2003). At the intersection of micro and macro: Opportunitiesand challenges for physician-patient communication research. PatientEducation and Counseling, 50(1), 13–16.

Cline, R.J.W., & Haynes, K.M. (2001). Consumer health information seek-ing on the Internet: The state of the art. Health Education Research,16(6), 671–692.

Cohen, A., Stotland, E., & Wolfe, D. (1955). An experimental investigationof need for cognition. Journal of Abnormal and Social Psychology, 51(2),291–294.


Coiro, J. (2003). Reading comprehension on the Internet: Expanding ourunderstanding of reading comprehension to encompass new literacies.Reading Teacher, 56(5), 458–464.

Coyle, J.R., & Thorson, E. (2001). The effects of progressive levels of in-teractivity and vividness in web marketing sites. Journal of Advertising,30(3), 65–81.

Detlefsen, E.G. (2004). Where am I to go? Use of the Internet for consumerhealth information by two vulnerable communities. Library Trends,53(2), 283–300.

Dias, P., Gomes, M.J., & Correia, A.P. (1999). Disorientation in hypermediaenvironments: Mechanisms to support navigation. Journal of EducationalComputing Research, 20(2), 93–117.

Dillon, A. (1991). Readers’ models of text structure: The case of academicmaterials. International Journal of Man-Machine Studies, 35(6), 913–925.

Dillon, A., & Gabbard, R. (1998). Hypermedia as an educational technology:A review of the quantitative research literature on learner comprehension,control, and style. Review of Educational Research, 68(3), 322–349.

Diseth, A., & Martinsen, O. (2003). Approaches to learning, cognitive style,and motives as predictors of academic achievement. Educational Psy-chology, 23(2), 195–207.

Downes, E.J., & McMillan, S.J. (2000). Defining interactivity: A qualitativeidentification of key dimensions. New Media & Society, 2(2), 157–179.

Evans, C., & Sabry, K. (2003). Evaluation of the interactivity of Web-basedlearning systems: Principles and process. Innovations in Education andTeaching International, 40(1), 89–99.

Eveland Jr., W.P., & Dunwoody, S. (2002). An investigation of elaborationand selective scanning as mediators of learning from the Web versusprint. Journal of Broadcasting & Electronic Media, 46(1), 34–53.

Eveland Jr., W.P., Cortese, J., Park, H., & Dunwoody, S. (2004). How Website organization influences free recall, factual knowledge, and knowledgestructure density. Human Communication Research, 30(2), 208–233.

Evers, K.E., Prochaska, J.M., Prochaska, J.O., Driskell, M.M., Cummins,C.O., & Velicer, W.F. (2003). Strengths and weaknesses of health behav-ior change programs on the Internet. Journal of Health Psychology, 8(1),63–70.

Fiore, A.M., & Jin, H.J. (2003). Influence of image interactivity on ap-proach responses towards an online retailer. Internet Research-ElectronicNetworking Applications and Policy, 13(1), 38–48.

Fox, S. (2005, May 15). Health information online (Pew Internet and Amer-ican Life Project Report). Retrieved October 19, 2005, from http://www.pewinternet.org/PPF/r/156/report_display.asp

Fox, S., & Fallows, D. (2003, July 16). Internet health sources: Healthsearches and e-mail have become more commonplace, but there is roomfor improvement in searches and overall Internet access (Pew Internet andAmerican Life Project Report). Retrieved October 5, 2003, from http://www.pewinternet.org/reports/pdfs/PIP_Health_Report_July_2003.pdf

Ha, L., & James, E.L. (1998). Interactivity reexamined: A baseline analysisof early business Web sites. Journal of Broadcasting & Electronic Media,42(4), 457–474.

Heeter, C. (2000). Interactivity in the context of designed experiences.Journal of Interactive Advertising, 1(1), 75–89.

Horrigan, J.B., & Rainie, L. (2002, December 29). Most expect to find keyinformation online, most find the information they seek, many now turnto the Internet first (Pew Internet and American Life Project Report). Re-trieved January 5, 2003, from http://www.pewinternet.org/reports/pdfs/PIP_Expectations.pdf

Jaffe, J.M. (1997). Media interactivity and self-efficacy: An examination ofhypermedia first aid instruction. Journal of Health Communication, 2(4),235–251.

Jensen, J.F. (1998). ‘Interactivity.’ Tracking a new concept in media andcommunication studies. Nordicom Review: Proceedings of the XIIINordic Conference on Mass Communication Research, 19(1), 185–204.

Jimison, H., Adler, L., Coye, M., Mulley Jr., A., & Eng, T.R. (1999). Healthcare providers and purchasers and evaluation of interactive health com-munication applications. American Journal of Preventive Medicine,16(1), 16–22.

Ketterer, S. (2001). The effects of links, story type, and personalityvariables on readers’ perceptions and use of crime stories in online

newspapers. Unpublished dissertation, University of Missouri,Columbia, MO.

Kiousis, S. (2002). Interactivity: A concept explication. New Media &Society, 4(3), 355–383.

Kovacs, D.K. (2004). Why develop Web-based health information work-shops for consumers? Library Trends, 53(2), 348–359.

Krane, D. (2005). Number of “cyberchondriacs”-U.S. adults who go onlinefor health information-increases to estimated 117 million. Harris Interac-tive Healthcare News, 5. Retrieved October 19, 2005, from http://www.harrisinteractive.com/news/newsletters/healthnews/HI_HealthCareNews2005Vol5_Iss08.pdf

Lawless, K.A., & Brown, S.W. (1997). Multimedia learning environments:Issues of learner control and navigation. Instructional Science, 25(2),117–131.

Lawless, K.A., & Kulikowich, J.M. (1994, February). Reading compre-hension, navigation, and hypertext. Paper presented at the AnnualMeeting of the Southwestern Educational Research Association, SanAntonio, Texas.

Lee, M.J. (2002). Effective tailored-communication in learning fromhypertext: Introducing expanding hypertext based on individuals’sensation-seeking and working memory capacity. Dissertation Ab-stracts International Section A: Humanities and Social Sciences,62(10A), 3224.

Lee, M.J., & Tedder, M.C. (2004). Introducing expanding hypertext basedon working memory capacity and the feeling of disorientation: Tailoredcommunication through effective hypertext design. Journal of Educa-tional Computing Research, 30(3), 171–195.

Leone, C., & Dalton, C.H. (1988). Some effects of the need for cognition oncourse grades. Perceptual and Motor Skills, 67(1), 175–178.

Levin, I.P., Huneke, M.E., & Jasper, J.D. (2000). Information processing atsuccessive stages of decision making: Need for cognition and inclusion-exclusion effects. Organizational Behavior and Human DecisionProcesses, 82(2), 171–193.

Light, A., & Wakeman, I. (2001). Beyond the interface: Users’ perceptionsof interaction and audience on websites. Interacting with Computers,13(3), 325–351.

Liu, Y.P. (2003). Developing a scale to measure the interactivity of Websites. Journal of Advertising Research, 43(2), 207–216.

Liu, Y.P., & Shrum, L.J. (2002). What is interactivity and is it always sucha good thing? Implications of definition, person, and situation for theinfluence of interactivity on advertising effectiveness. Journal of Adver-tising, 31(4), 53–64.

MacGregor, S.K. (1999). Hypermedia navigation profiles: Cognitive char-acteristics and information processing strategies. Journal of EducationalComputing Research, 20(2), 189–206.

Martindale, C. (1991). Cognitive psychology: A neural-network approach.Pacific Grove, CA: Brooks/Cole.

Massey, B.L., & Levy, M.R. (1999). Interactivity, online journalism andEnglish-language web newspapers in Asia. Journalism & Mass Commu-nication Quarterly, 76(1), 138–151.

McDonald, S., & Stevenson, R.J. (1996). Disorientation in hypertext: Theeffects of three text structures on navigation performance. AppliedErgonomics, 27(1), 61–68.

McDonald, S., & Stevenson, R.J. (1998). Navigation in hyperspace: Anevaluation of the effects of navigational tools and subject matter exper-tise on browsing and information retrieval in hypertext. Interacting withComputers, 10(2), 129–142.

McMillan, S.J. (1999). Health communication and the Internet: Relationsbetween interactive characteristics of the medium and site creators, con-tent, and purpose. Health Communication, 11(4B), 375–391.

McMillan, S.J. (2000). Interactivity is in the eye of the beholder: Function,perception, involvement, and attitude toward the web site. Paper pre-sented at the 2000 Conference of the American Academy of Advertising,Michigan State University, East Lansing, MI.

McMillan, S.J., & Hwang, J.S. (2002). Measures of perceived interactivity:An exploration of the role of direction of communication, user control,and time in shaping perceptions of interactivity. Journal of Advertising,31(3), 29–42.


Narayanan, N.H., & Hegarty, M. (2002). Multimedia design for comm-unication of dynamic information. International Journal of Human-Computer Studies, 57(4), 279–315.

Nelson, W.A., & Palumbo, D.B. (1992). Learning, instruction, and hyper-media. Journal of Educational Multimedia and Hypermedia, 1(3),287–299.

Newhagen, J.E., & Cordes, J.W. (1995). [email protected]: Audiencescope and the perception of interactivity in viewer mail on the Internet.Journal of Communication, 45(3), 164–176.

Newhagen, J.E., & Rafaeli, S. (1996). Why communication researchersshould study the Internet: A dialogue. Journal of Communication, 46(1), 4.

Niederhauser, D.S., Reynolds, R.E., Salmen, D.J., & Skolmoski, P. (2000).The influence of cognitive load on learning from hypertext. Journal ofEducational Computing Research, 23(3), 237–255.

Pavlou, P.A., & Stewart, D.W. (2000). Measuring the effects and effective-ness of interactive advertising: A research agenda. Journal of InteractiveAdvertising, 1(1). Retrieved October 19, 2005, from http://jiad.org/vol1/no1/pavlou/index.htm

Petty, R.E., & Cacioppo, J.T. (1979). Issue involvement can increase or de-crease persuasion by enhancing message-relevant cognitive responses.Journal of Personality and Social Psychology, 37(10), 1915–1926.

Petty, R.E., & Cacioppo, J.T. (1986). The elaboration likelihood model ofpersuasion. Advances in Experimental Psychology, 19, 123–205.

Petty, R.E., Cacioppo, J.T., & Goldman, R. (1981). Personal involvement asa determinant of argument-based persuasion. Journal of Personality andSocial Psychology, 41(5), 847–855.

Piaget, J. (1970). Science of education and the psychology of the child.New York: Orion Press.

Rafaeli, S., & Sudweeks, F. (1997, March). Networked interactivity. Journalof Computer-Mediated Communication, 2. Retrieved December 4, 2003,from the World Wide Web: http://www.ascusc.org/jcmc/vol2/issue4/rafaeli.sudweeks.html

Reeves, B., & Nass, C. (2000). Perceptual bandwidth. Communications ofthe ACM, 43(3), 65–71.

Rice, R.E., & Katz, J.E. (2001). The Internet and health communication:experiences and expectations. Thousand Oaks, CA: Sage Publications.

Salomon, G. (1979). Interaction of media, cognition, and learning (1st ed.).San Francisco: Jossey-Bass Publishers.

Schacter, J., & Fagnano, C. (1999). Does computer technology improvestudent learning and achievement? How, when, and under what condi-tions? Journal of Educational Computing Research, 20(4), 329–343.

Smith, S.M., & Levin, I.P. (1996). Need for cognition and choice framingeffects. Journal of Behavioral Decision Making, 9(4), 283–290.

Spiro, R.J., & Jheng, J. (1990). Cognitive flexibility and hypertext: Theoryand technology for the non-linear and multidimensional traversal of

complex subject matter. In R.J. Spiro (Ed.), Cognition, education, andmultimedia: Exploring ideas in high technology (p. 165). Hillsdale, NJ:Lawrence Erlbaum Associates.

Steuer, J. (1992). Defining virtual reality: Dimensions determining telepres-ence. Journal of Communication, 42(4), 73–93.

Stout, P.A., Villegas, J., & Kim, H. (2001). Enhancing learning through useof interactive tools on health-related Web sites. Health Education Re-search, 16(6), 721–733.

Street, R.L., & Rimal, R.N. (1997). Health promotion and interactivetechnology: A conceptual foundation. In R.L. Street, W.R. Gold, &T. Manning (Eds.), Health Promotion and Interactive Technology: Theo-retical Applications and Future Directions (pp. 1–18). Mahwah, NJ:Lawrence Erlbaum Associates.

Sundar, S.S., Kalyanaraman, S., & Brown, J. (2003). Explicating Web siteinteractivity. Communication Research, 30(1), 30–59.

Sundar, S.S., Narayanan, S., Obregon, R., & Uppal, C. (1998). Does Webadvertising work? Memory for print vs. online media. Journalism andMass Communication Quarterly, 75, 822–835.

Swinyard, W.R., & Smith, S.M. (2003). Why people (don’t) shop online: Alifestyle study of the Internet consumer. Psychology and Marketing,20(7), 567.

Tewksbury, D., & Althaus, S. (2000). Differences in knowledge acquisitionamong readers of the paper and online versions of a national newspaper.Journalism and Mass Communication Quarterly, 77, 477–479.

Tremayne, M., & Dunwoody, S. (2001). Interactivity, information process-ing, and learning on the World Wide Web. Science Communication,23(2), 111–134.

Tuten, T.L., & Bosnjak, M. (2001). Understanding differences in Webusage: The role of need for cognition and the five factor model of per-sonality. Social Behavior and Personality, 29(4), 391–398.

Unz, D.C., & Hesse, F.W. (1999). The use of hypertext for learning. Journalof Educational Computing Research, 20(3), 279–295.

Venkatraman, M.P., Marlino, D., Kardes, F.R., & Sklar, K.B. (1990). Theinteractive effects of message appeal and individual differences on infor-mation processing and persuasion. Psychology and Marketing, 7(2),85–96.

Verplanken, B., Hazenberg, P.T., & Palenewen, G.R. (1992). Need for cog-nition and external information search effort. Journal of Research in Per-sonality, 26(2), 128–136.

Waniek, J., Brunstein, A., Naumann, A., & Krems, J.F. (2003). Interactionbetween text structure representation and situation model in hypertextreading. Swiss Journal of Psychology, 62(2), 103–111.

Zhang, Z., Han, M., Zhu, Z., & Zhu, W. (2002). Effects of text structure andtime stress on Web reading performance. Psychological Science China,25(4), 422–424.

can interactivity make a difference

Documents