The Persuasion Context and Results in Online Opinion Seeking: Effects of Message and Source

Abstract

Online opinion networks are areas for social exchange, or conversational networks, made up of individuals actively involved in sharing experiences and opinions concerning matters of mutual interest between consumers or concerning their experience with a given product or service. We pinpoint a gap in the literature regarding how the persuasion process occurs when individuals seek opinions online, including the results process. In an attempt to find an answer, we draw on traditional theories related to information processing. These are mostly taken from the field of psychology and enable us to identify which signals or aspects of communication or opinions the individuals focus their attention on (message and source) and the value attached to such communications as well as how much they impact individuals' purchase decisions, bearing in mind the medium (or online opinion network) in which the opinions are located. Findings from those interviewed support the idea that the quality of information on the Internet, as well as trust in the source of said information, or in the opinion of network users, have an impact on the informational value obtained from involvement in this online opinion seeking and on purchasing decisions. Moreover, depending on the kind of network (firm or brand controlled, review Web sites, and user-controlled nonofficial opinion networks), the quality of the information or trust in the users will have a different bearing in the persuasion process.

Introduction

Online interpersonal influence or electronic word of mouth (WOM) is an important aspect of the Internet.¹ Unlike conventional means, where information is passed on taxonomically or selectively, in the new 2.0 environment information is distributed through the Net and is based on folksonomic or collaborative principles. In other words, members of a network are involved and contribute toward generating and distributing information.^2,3

Online opinion networks are areas for social exchange, or conversational networks, made up of individuals actively involved in sharing experiences and opinions concerning matters of mutual interest.^4–6 Applied to a marketing setting, WOM refers to a conversation between consumers concerning their experience with a given product or service.⁷ The unique asynchronous and interactive nature of cyberspace provides consumers with unparalleled access to information, wide product and brand choice, the ability to make price and quality comparisons as never before, and the chance to interact with companies and other consumers in a variety of ways.⁸ These interactions are conducted via Web pages, blogs, forums, review sites, and social networking sites, among others.⁹

Numerous studies have explored the motivations that lead individuals to seek the opinions of others on the Internet¹ and share them,^10–12 whereas other studies have explored the consequences thereof.^13–15 Yet, we pinpoint a gap in the literature regarding how the persuasion process occurs when individuals seek opinions online, including the results process. This is our main interest of research.

The persuasion process in the search for online opinions

Since the pioneering work of Twedt¹⁶ and Diamond,¹⁷ numerous authors have focused their attention on how communication execution variables impact their efficiency,^18–20 also in the online medium.^21–24 In most cases a distinction is made between variables related to content or message and variables related to the execution or external configuration of the communication (i.e., music, humor). Following the elaboration likelihood model (ELM), developed by Petty et al.¹⁸ and Petty and Cacioppo,²⁵ we distinguish between the importance of the arguments or information quality of the message, which act as central cues, and the characteristics of the source (credibility, source expertise, or trust in the source, among others), which act as peripheral cues^26–28 in creating individuals' attitude and behavior (or persuasion process). Although the authors¹⁸ of the ELM have traditionally maintained that the persuasion process is established on a continuum between the central and peripheral processing routes (corresponding to situations of high and low involvement, or to cognitive and affective involvement, respectively), which are determinants of the kind of cues that individuals focus on, several authors have shown that in an online context the two routes may interact and work together to influence the online persuasion process.^24,29 This is particularly true of online opinions spaces, where information is exchanged concerning various kinds of products for which individuals display both differing levels as well as different types of involvement. For instance, in review Web sites, the same person may seek information about financial products, generally linked to cognitive or rational involvement; fashion, usually more closely related to involvement of an affective or emotional nature; or travel, generally linked to both cognitive and affective involvement at the same time, among many other possibilities.

As a result, when individuals seek others' opinions on the Internet as a source of information in their purchase decision-making process, both the source (as a peripheral cue that is particularly valued in low-involvement situations or those related to more affective or emotional involvement) and the value of the message or the information sought (as a central cue) are involved. The information obtained by individuals after interacting in these spaces (or the perceived informative value of involvement in the online opinions space) will help them to take a decision and will make them feel more confident when purchasing. It has been evidenced that the trust placed in the users who make up a network determines the satisfaction of whoever is seeking the online opinions.^29,30,31 Furthermore, the quality of the knowledge shared (in our case, the information quality of the online opinions space), also impacts said satisfaction.³² In fact, the informational value that consumers receive from online information lies precisely in the fact that it aids the evaluation and decision-making process.^33,34 We therefore propose the following:

H1: The information quality of online opinion networks affects individuals' informational value in their purchasing process.

H2: Trust in online opinion network users affects individuals' informational value in their purchasing process.

As in conventional media, we believe that the informational value generated when an individual seeks information during a purchase process influences the behavior purchase of new products or brands, how much is spent, and so on. According to Goldsmith and Horowitz,¹ in online opinion networks, consumers seek the opinions of others online to reduce risk, because others do so, to secure lower prices, to obtain information easily, because it is cool, and to obtain prepurchase information, among other reasons—the influence of other consumers' information proving to be more important than advertising. Several authors express serious doubts as to whether online opinions are actually effective because of the anonymity among communicators³⁵ or the absence of interpersonal physical or face-to-face contact among network members.³⁶ As a result, given the particular nature of this new context in which communication takes place, there is a need to explore to what extent the informational value gained from involvement in online opinion networks impacts individuals' attitudes and behavior, in short, their purchase behavior.

H3: The informational value individuals' gain from participation in online opinion networks affects their purchase behavior.

The moderating role of the kind of online opinion network

As suggested by Johnson and Kaye,³² we felt that it would prove enlightening to analyze the moderating effect of the kind of online opinion network on the impact of the message content and to gauge the impact of the source on the purchase behavior of individuals who consult online opinions. According to Brown et al.,²⁹ social network theories hold that individual behavior, group behavior, and organizational behavior are affected more by the kinds of ties and networks in which the actors are involved than by the individual attributes of the actors themselves.³⁷ We distinguish three types of online opinion network depending on who is managing it: the firm or product brand about which we are seeking opinions (through forums or blogs or in official profiles in social networks), a firm or Web site specialized in gathering opinions on different products or online brands (review sites or recommended systems, such as Ciao), or nonofficial spaces run by other anonymous users in their blogs or in their social network profiles. Bearing in mind that the medium in which the advertising is placed determines the kind of processing to occur during exposure to communication,³⁸ our aim is to explore how this impacts processing of central and peripheral cues (information quality and trust in the source) in our study—in other words, how the type of network impacts the individual persuasion process. We believe that the moderating effect will be reflected in the following manner. When accessing an official space run by the product or brand, or by a recognized firm specializing in opinions, individuals seeking opinions about a product are expected to have a predetermined attitude toward said network (either through knowledge thereof or previous experience with the product, brand, or said review site). Said predisposition and voluntary access reflects high involvement behavior toward the firm handling the opinions, and, in line with Petty et al.,¹⁸ processing is likely to take place through a central route. In other words, persuasion is a result of the informative value of the opinions expressed concerning this product. By contrast, when individuals access networks run by unknown users and about which they have no previous experience or preconceived opinion, processing occurs through the peripheral route. In other words, attention focuses on the peripheral elements of communication (trust in users or source of the opinions). Given that the persuasion process takes place in a continuum between the central and peripheral routes for processing,^24,29 it is to be expected that:

H4: In official (or nonuser-controlled) online opinion networks, the information quality might prove more important, whereas for the nonofficial (or user-controlled) opinion networks, trust in the source will prove more important in the persuasion process.

Figure 1 provides a global representation of the relationships we seek to explore.

FIG. 1.

Proposed model.

Materials and Methods

Information gathering and characteristics of the sample

Information was gathered through a questionnaire given to users of forums, blogs, social networks, and other online spaces that bring together opinions from consumers and users. The questionnaire was handed out personally, and an online version was also drawn up and posted with a hyperlink in certain forums and blogs to encourage more active participants to complete it. Once invalid questionnaires had been removed, the total sample came to 309.

Subjects were asked to choose a network (including forums, blogs, social networks, or Web sites specialized in gathering consumers' opinions or remarks) in which they were involved or which they visited at least occasionally, and were asked to indicate who ran the space (the brand, a firm specialized in gathering opinions, or private users). Such a process enabled us to gather information from communities covering a wide range of topics. Table 1 details the characteristics of the sample.

Table 1.

Characteristics of the Sample

Type of online opinion network^a (%)	Topics^b (%)	Sex (%)	Age^c (%)
Firm or brand-controlled (47.9)	Travel (37.9)	Female (55.2)	Under 18 (6.9)
Review Web sites (28.6)	IT and electronics (34.6)	Male (44.8)	Between 19 and 25 (76.4)
User-controlled or nonofficial (11.4)	Music (34.3)		Between 26 and 35 (6.2)
Others (12.1)	Fashion (31.7)		Between 36 and 45 (4.3)
	Cinema and shows (28.8)		Between 46 and 55 (5.2)
	Sport (24.3)		Above 55 (1.0)
	Books (12.0)
	Videogames (12.0)
	Food (11.7)
	Financial products (6.8)
	Others^d (16.8)

Of the 309 individuals surveyed, in 19 cases we obtained no response regarding the type of community (missing values). As a result, these percentages are calculated on the basis of 290 observations.

In many cases, subjects cited forums or Web sites that covered a variety of topics.

Age distribution is comparable to that used by Wu et al.³¹

Cars, cosmetics, photography, the home, and so on.

Measurement of variables

The information quality scale is adapted on several works that measure the quality of the information online contexts.^39–41 The trust scale is Chiu et al's³⁰ and is created based on the traditional Doney and Cannon⁴² trust scale. Informational value and influence on purchasing behavior are based on the co-shopping scale of Chan and Li,³³ but taking into account the particular features of the phenomenon being explored. On the basis of in-depth interviews and pretests conducted, we adapted and added fresh items, opting to distinguish the process (informational value) from the behavior (purchasing behavior). Specifically, to measure information quality, we use items closely connected with the features of the information circulating in the opinions network (extent, relevant, easy to understand, accurate and detailed, comprehensive, timely and reliable), whereas the informative value reflects individuals' perception of how their participation in said space helps them to make better purchase decisions and to do so more easily and with greater confidence, and thus leads them to refer to this opinions space before purchasing.

To measure the variables proposed in the study, in most cases we employed scales used and validated in previous studies.^10,31,33 To validate these scales, we previously conducted confirmatory factor analysis over the whole sample. The goodness of fit indicators may be considered acceptable. We subsequently estimated the proposed model using partial least squares (PLS) analysis.^a Table 2 shows the indicators and their respective descriptive statistics, as well as the standardized factor loadings of the proposed measurement model resulting from the PLS analysis. Also shown are Cronbach's alpha coefficient values, composite reliability, and average variance extracted (AVE) of each construct measured with several indicators.

Table 2.

Variables, Descriptive Statistics, and Reliability and Validity Tests

Constructs and indicators	Mean	SD	λ_i (CFA ^a )	λ_i (PLS)
Information Quality (α=0.84; CR=0.88; AVE=0.51)
This online opinion network offers a great amount of information.	3.67	1.00	0.64	0.78
The topics covered are relevant.	3.41	0.99	0.65	0.74
The information provided is easy to understand.	3.77	0.85	0.69	0.72
The information provided is accurate and detailed.	3.47	0.90	0.76	0.76
The information provided is complete.	3.39	0.93	0.72	0.75
The information is timely.	3.37	0.94	0.63	0.60
The information is truthful.	3.11	1.02	0.59	0.62
Trust in Online Opinion Network Users (α=0.88; CR=0.92; AVE=0.74)Participants in this online opinion network…
Would not take advantage of others even when the opportunity arises.	2.58	1.01	0.65	0.79
Always keep their promises.	2.32	0.87	0.84	0.88
Behave in a consistent and honest manner.	2.76	0.98	0.87	0.89
Are truthful in dealing with one another.	2.85	0.96	0.83	0.87
Informational Value (α=0.91; CR=0.94; AVE=0.79)As a result of my participation in this online opinion network…
I consult this online opinion network before buying.	3.11	1.23	0.79	0.85
I make better purchase decisions.	3.15	1.10	0.90	0.90
Making a purchase decision proves easier.	3.24	1.09	0.92	0.92
I feel more confident when I buy.	3.08	1.14	0.84	0.87
Influence on Purchase Behavior (α=0.90; CR=0.93; AVE=0.77)As a result of my participation in this online opinion network…
I make more purchases online than before.	2.68	1.34	0.80	0.87
I buy products or services that I did not buy before.	2.57	1.25	0.80	0.86
I buy more products of certain brands than before.	2.49	1.20	0.81	0.89
I spend a larger amount of money on this type of products.	2.51	1.25	0.88	0.89
Frequency of Use^b (α=0.79; CR=0.91; AVE=0.83)
Frequency of access to the online opinion network	3.49	1.28	0.85	0.93
Frequency of active participation in the online opinion network	2.81	1.33	0.86	0.89
Number of Followers^c (n.a.)
Number of followers of the online opinion network	3.85	1.16		1.00

Goodness of fit: χ²(179)=359.71 (p=0.000); χ²/df=2.01; GFI=0.898; AGFI=0.868; CFI=0.951; RMR=0.056; RMSEA=0.057.

Frequency of use was measured with two indicators reflecting how often the respondent accesses to site and how often he/she reads and makes comments, where 1 meant never; 2, sporadically; 3, once a week; 4, several times a week; 5, daily.

The number of followers was measured with a single item in which 1 means that the reported community had fewer than 100 followers, 2 means between 100 and 1000 followers, 3 means between 1000 and 10000, 4 means between 10000 and 100000, and 5 means over 100000 followers.

α, Cronbach's alpha; AGFI, adjusted goodness of fit index; AVE, average variance extracted; CFA, confirmatory factor analysis; CFI, comparative fit index; CR, composite reliability; GFI, goodness of fit index; n.a., not applicable; PLS, partial least squares; RMR, root mean residual; RMSEA, root-mean-squared error of approximation.

As control variables, we included the frequency of use of the Web community and its number of followers in order to assess to what extent the significance or lack of significance of the effects proposed in our hypotheses might be explained by frequency or regularity of subject participation and by the popularity of the community and not so much by the relationships theoretically posited.

All the standardized item loadings are significant and are, in general, above the recommended value of 0.7, all of them indeed being greater than 0.6. For all the constructs, Cronbach's alpha coefficient and composite reliability index values are above 0.7, and AVE is well above the recommended minimum value of 0.5, allowing us to conclude that our scale displays sufficient convergent validity.

Table 3 shows the means, standard deviations, and correlations among the constructs for all the variables included in our study. In line with MacKenzie et al.,⁴⁵ for each construct we verified that the square root of its AVE is greater than its correlation with the remaining constructs and that all the correlations are below 0.71.

Table 3.

Construct Correlations

Construct	1	2	3	4	5	6
1. Information quality	0.71 ^a
2. Trust in users	0.33	0.86
3. Informational value	0.39	0.25	0.89
4. Influence on purchase behavior	0.21	0.17	0.58	0.88
5. Frequency of use	−0.05	0.09	−0.06	0.08	0.91
6. No. of followers	0.04	−0.12	0.05	0.05	0.26	n.a.

Diagonal elements (italic) show the square root of the average variance extracted.

n.a., not applicable.

Results

We estimated our model using the SmartPLS (version 2.0.M3) software developed by Ringle et al.⁴⁶ The level of statistical significance of the coefficients of both the measurement and the structural models was determined through a bootstrap resampling procedure (500 subsamples were randomly generated).

As can be seen in Table 4, the proposed hypotheses are supported. First, in line with hypotheses H1, H2, and H3, and after controlling for the frequency of use and the number of followers, we find that the influence on purchasing process is positively related to the quality of the information obtained therein (H1, β=0.33, p<0.01) and trust in network users (H2, β=0.21, p<0.01), and that the purchasing process influences purchase behavior (H3, β=0.62, p<0.01)—that is, participants acknowledge that they make a greater amount of online purchases, buy new products and services or consume more products of certain brands, and spend more money on the products recommended in the community. Furthermore, information quality and trust in users have a significant and positive indirect (via purchasing process) influence on purchase behavior.

Table 4.

Partial Least Squares Results: Standardized Parameter Estimates

		Type of community
	Complete sample (n=255)^a	Controlled by firm/brand (n=139)	Review site/multibrand (n=83)	Controlled by the users (n=33)
Hypothesized relationships
Info. quality→Informational value	0.33^**	0.31^**	0.54^**	0.05
Trust in users→Informational value	0.15^*	0.13	0.05	0.72^**
Influence of purchasing process→Influence on purchase behavior	0.59^**	0.65^**	0.48^**	0.76^**
Control relationships
Frequency of use→Informational value	−0.08	0.00	−0.18	0.12
Frequency of use→Influence on purchase behavior	0.12^*	0.04	0.10	0.24^*
No. of followers→Influence of purchasing process	0.08	0.03	0.08	0.24
No. of followers→Influence on purchase behavior	−0.01	0.05	0.02	−0.30
R² of Informational value	0.177	0.138	0.382	0.480
R² of Influence on purchase behavior	0.353	0.424	0.225	0.660
Indirect effects
Info. quality→Influence on purchase behavior	0.20^**	0.20^**	0.26^**	0.04
Trust in users→Influence on purchase behavior	0.13^**	0.08	0.02	0.55

Significance levels are determined using bootstrapping.

The study also included the option of other types of networks. Because of the impossibility of recoding them, we opted to remove them in order to carry out the corresponding statistical analysis. We also omitted those cases with missing values in this variable.

p<0.05, ^**p<0.01 (one-tailed test for hypothesized relationships, two-tailed test for control relationships).

Hypothesis H4 concerning the moderating effect of the owner or person running the network, or type of network, on the relationships proposed is also supported by our data. We contended that the relative importance of information quality and trust in the source varies depending upon the type of virtual community considered. We tested this hypothesis by conducting a PLS-based multigroup analysis.³⁹ We estimated the model for the three types of networks identified in our study and assessed how the significance and the magnitude of the path parameters vary across types. For networks controlled or sponsored by a firm or brand, the quality of the information provided is the major determinant in users' informational value in their purchasing process (β=0.31, p<0.01) as a source of information and a means for making more accurate, easier, and safer buying decisions, while trust in other users seems to be insignificant (β=0.13, p>0.05). This pattern is very similar to that found for review sites; that is, for networks of this kind, the purchasing process is mainly driven by the quality of the information they provide (β=0.54, p<0.01) and not by the perceived trust in other users (β=0.05, p>0.05). In clear contrast with these findings, for user-controlled networks and nonofficial profiles of a firm or brand in social networks, we observe a significant and very strong effect of trust in network users on the informational value (β=0.72, p<0.01), while the effect on satisfaction of information quality is not significant (β=0.05, p>0.05).

To further demonstrate that the type of network moderates the relationships proposed in H1 and H2, we examined whether the corresponding path coefficients differ across subsamples using Keil et al.'s⁴⁷ approach. Thus, we observe that the effect of trust in network users on purchasing process is significantly greater (β⁽¹⁾−β⁽²⁾=0.59, p<0.01) for user-controlled communities (β=0.72) than for firm or brand-controlled communities (β=0.13), and that the effect of information quality on purchasing process is greater (although only marginally; β⁽¹⁾ – β⁽²⁾=0.26, p<0.07) for the latter type of network (β=0.31) than for the former (β=0.05). We likewise observe that the effect of information quality on purchasing process is statistically greater (β⁽¹⁾−β⁽²⁾=0.23, p<0.05) for review sites (β=0.54) than for firm or brand-controlled networks (β=0.31).

Discussion

Online opinion networks are frequently consulted by users when making certain purchase decisions. We analyze to what extent the influence these sites have on subjects' purchases depends on how much individuals trust the information provided or the users involved.

The study carried out proves that the message (informative quality) and the source (trust in users) do actually influence informational value subjects' in purchase process (they consult opinions more often, take purchase decisions more easily, and feel more sure when they do buy) and, as a result, also impact their purchase behavior (buying new products or brands, amount spent, online purchases, etc.). The most noticeable finding relates to the moderating effect of the kind of network depending on who is running it (the brand, a specialized firm, or an anonymous user), on the execution variable, or the kind of cue most strongly impacting user purchase behavior. When the information consulted by consumers comes from an official brand site or a review Web site, the influence on informational value on the process and on subsequent purchase decisions depends on the amount of trust placed in the quality of the information and not on the trust placed in the users involved in the network. This finding indicates that trust in the brand or in the firm does not imply that consumers trust the online information these offer or that they use it when making decisions. Only when subjects perceive that the information being provided by the firm's Web site is useful, relevant, understandable, comprehensive, detailed, and timely will they be willing to use it in their purchase decisions. The opposite is true for Web sites controlled or run by a user (nonofficial blogs and profiles in social networks). In these cases, subjects base their reaction on the trust they place in the source of information, like a peripheral cue, which allows them to trust the Web site and consequently make purchase decisions.

Management implications

Although many firms have opted for Web socialization by including corporate forums and blogs in their Web sites and by becoming involved in social networks, our study has shown that these tools only influence consumers if the contents posted actually prove useful and relevant to the user. On its own, trust in the brand or institution is not enough to make individuals value the information and use it in their decisions.

Firms do, however, have other alternatives. It has been shown that consumers value blogs and social networks run by users because of the trust conveyed by those who are involved in them or who manage them. In this vein, Johnson and Kaye³² point out that the credibility of blogs lies in the fact that they contain information that cannot be found in other conventional media and that the topics dealt with are discussed in greater depth. When individuals trust the information source, they expect it to provide not just a greater amount of information but also more detailed and better prepared information than can be found in other conventional media. Firms should therefore find out how to cooperate with these online opinion leaders in order to make them the intermediaries in communication with clients.

With regard to multibrand discussion Web sites run by third parties who specialize in opinion gathering, we have also noticed that when there is a “faceless crowd” behind an opinion Web site, individuals will only value the site depending on the quality of the content. These community managers must include indications (evaluations, votes, etc.) that allow the interest, relevance, or quality of the comments posted to be specified. In addition, rating each source of information (or content producer) would enable said sources to be ranked depending on their credibility and would therefore facilitate consumers' search for information.

Limitations and further research

The study's main limitations are that it is the individuals surveyed who subjectively classify the kind of network they analyze in their responses, and that the size of certain subsamples, particularly regarding spaces controlled by users, is small. This problem means that the estimators obtained might be biased and therefore restrict the work's statistical power,⁴³ although it has not prevented us from pinpointing statistically significant effects not only for the sample as a whole but also for multigroup analysis.

Moreover, the present study focuses on exploring individuals' subjective valuation of the impact that their participation in online opinions spaces has on their purchase behavior with regard to intentions. Testing the model in future using effective behavior data is no doubt recommendable to ensure the model's predictive capacity. Following a different line, we intend to explore in greater depth the features of the individuals exposed to these spaces, such as their prior attitudes toward said spaces. We will also seek to gauge to what extent these characteristics might impact their valuations and the findings to emerge from our study.

As regards the characteristics of online opinions spaces, future studies might examine what kind of cues prove most credible for consumers and help them in their search for information. We also feel that future research should address the informative content of social networks. The fact that the information posted on these types of media ages rapidly is a factor to be taken into account when estimating the effectiveness of communication therein.

Acknowledgment

The authors gratefully acknowledge financial support from the Regional Ministry for Education (VA001B10–1) at the Regional Government (Castilla y León, Spain).

Notes

a. As mentioned above, PLS path modeling was chosen as the analytical technique to test the research hypotheses mainly because, according to Chin and Newsted⁴³ and Henseler et al.,⁴⁴ among other advantages, PLS does not impose any distributional assumption for the observed variables and because it enables path models to be estimated when sample sizes are small (such as our case, particularly when estimating the model for user-controlled web communities).

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Goldsmith

, Horowitz

. Measuring motivations for online opinion seeking. Journal of Interactive Advertising, 2006; 6:1–16.

Hennig-Thurau

, Gwinner

, Walsh

, Gremler

. Electronic word-of-mouth via consumer-opinion platforms: what motivates consumers to articulate themselves on the Internet? Journal of Interactive Marketing, 2004; 18:38–52.

Chiang

, Huang

. Traffics metrics and Web 2.0-ness. Online Information Review, 2010; 34:115–126.

Lee

FSL

, Vogel

, Limayem

. Virtual community informatics: a review and research agenda. Journal of Information Technology Theory and Application, 2003; 5:47–61.

Wiertz

, Ruyter

. Beyond the call of duty: why customers participate in firm-hosted commercial online communities. Organization Studies, 2007; 28:347–376.

, Zhao

, Wang

. From virtual community members to C2C e-commerce buyers: trust in virtual communities and its effect on consumers' purchase intention. Electronic Commerce Research and Applications, 2010; 9:346–360.

East

. The impact of positive and negative word of mouth on brand choice. ANZMAC 2005 Conference: Advertising/Marketing Communication Issues http://anzmac.info/conference/2005/cd-site/pdfs/1-Advertising/1-East.pdf. 2013 May 8.

Negroponte

, Maes

. Electronic word of mouth. Wired, 19964.10October.

Goldsmith

. 2006. Electronic word-of-mouth. Khosrow-Pour

. Encyclopedia of E-Commerce, E-Government and Mobile Commerce. Chapter 67:408–412. Hershey, PA: Idea Group Publishing.

10.

Hsu

, Ju

, Yen

, Chang

. Knowledge sharing behavior in virtual communities: the relationship between trust, self-efficacy, and outcome expectations. International Journal of Human-Computer Studies, 2007; 65:153–169.

11.

Trusov

, Bucklin

, Pauwels

. Effects of word-of-mouth versus traditional marketing: findings from an Internet social networking site. Journal of Marketing, 2009; 73:90–102.

12.

Wodzicki

, Schwämmlein

, Cress

, Kimmerle

. Does the type of anonymity matter? The impact of visualization on information sharing in online groups. Cyberpsychology, Behavior, and Social Networking, 2011; 14:157–160.

13.

Smith

, Coyle

, Lightfoot

, Scott

. Reconsidering models of influence: the relationship between consumer social networks and word-of-mouth effectiveness. Journal of Advertising Research, 2007; 47:387–397.

14.

Huang

, Lin

. Factors affecting pass-along email intentions (PAEIs): integrating the social capital and social cognition theories. Electronic Commerce Research and Applications, 2009; 8:160–169.

15.

Gupta

, Harris

. How e-WOM recommendations influence product consideration and quality of choice: a motivation to process information perspective. Journal of Business Research, 2009; 63:1041–1049.

16.

Twedt

. A multiple factor analysis of advertising readership. Journal of Applied Psychology, 1952; 36:207–215.

17.

Diamond

. A quantitative approach to magazine advertisement format selection. Journal of Advertising Research, 1968; 5:376–386.

18.

Petty

, Cacioppo

, Schumann

. Central and peripheral routes to advertising effectiveness: the moderating role of involvement. Journal of Consumer Research, 1983; 10:135–146.

19.

MacKenzie

, Lutz

, Belch

. The role of attitude toward the ad as a mediator of advertising effectiveness: a test of competing explanations. Journal of Marketing Research, 1986; 23:130–143.

20.

MacInnis

, Jaworski

. Information processing from advertisements: toward an integrative framework. Journal of Marketing, 1989; 53:1–23.

21.

Cho

. How advertising works on the WWW: modified elaboration likelihood model. Journal of Current Issues and Research in Advertising, 1999; 21:33–50.

22.

Bellman

, Rossiter

. The website schema. Journal of Interactive Advertising, 2004; 4:2.

23.

Karson

, Fisher

. Reexaminating and extending the dual mediation hypothesis in an on-line advertising context. Psychology & Marketing, 2005; 22:333–351.

24.

SanJosé-Cabezudo

, Gutiérrez-Arranz

, Gutiérrez-Cillán

. The combined influence of central and peripheral routes in the online persuasion process. Cyberpsychology, Behavior, and Social Networking, 2009; 12:299–308.

25.

Petty

, Cacioppo

. 1986. Communication and persuasion. Central and peripheral routes to attitude change. New York: Springer-Verlag.

26.

Petty

, Cacioppo

. Issue involvement as a moderator of the effects on attitude of advertising content and context. Advances in Consumer Research, 1981; 8:20–24.

27.

Petty

, Cacioppo

, Goldman

. Personal involvement as a determinant of argument based persuasion. Journal of Personality and Social Psychology, 1981; 41:847–855.

28.

Petty

, Cacioppo

, Heesacker

. Effects of rhetorical questions in persuasion: a cognitive response analysis. Journal of Personality and Social Psychology, 1981; 40:432–440.

29.

Brown

, Broderick

, Lee

. Word of mouth communication within online communities: conceptualizing the online social network. Journal of Interactive Marketing, 2007; 21:2–20.

30.

Chiu

, Hsu

, Wang

ETG

. Understanding knowledge sharing in virtual communities: an integration of social capital and social cognitive theories. Decision Support Systems, 2006; 42:1872–1888.

31.

, Chen

, Chung

. Trust factors influencing virtual community members: a study of transaction communities. Journal of Business Research, 2010; 63:1025–1032.

32.

Johnson

, Kaye

. In blog we trust? Deciphering credibility of components of internet among politically interested internet users. Computers in Human Behavior, 2009; 25:175–182.

33.

Chan

, Li

. Understanding consumer-to-consumer interactions in virtual communities: the salience of reciprocity. Journal of Business Research, 2010; 63:1033–1040.

34.

Huang

, Chouy

, Lin

. Involvement theory in constructing bloggers' intention to purchase travel products. Tourism Management, 2010; 31:513–526.

35.

Guadagno

, Cialdini

. 2005. Online persuasion and compliance: social influence on the Internet and beyond. Amichai-Hambuerger

. The Social Net: understanding human behavior in cyberspace. New York: Oxford University Press, 91–114.

36.

Gershoff

, Broniarkrczyk

, West

. Recommendation or evaluation? Task sensitivity in information source selection. Journal of Consumer Research, 2011; 28:418–438.

37.

Haythornthwaite

. 1999. A social network theory of tie strength and media use: a framework for evaluating multilevel impacts of new media. Technical Report UIUCLIS—2002/1+DKRC, Graduate School of Library and Information Science. Champaign, IL: University of Illinois at Urbana-Champaign.

38.

Krugman

. The impact of television advertising: learning without involvement. Public Opinion Quarterly, 1965; 29:349–356.

39.

Lin

. The role of online and offline features in sustaining virtual communities: an empirical study. Internet Research, 2007; 17:119–138.

40.

Kuo

, Lee

. KMS adoption: the effects of information quality. Management Decision, 2009; 47:1633–1651.

41.

Wang

, Sy

, Fang

. The post-adoption behavior of online knowledge community: decomposing customer value. The Journal of Computer Information Systems, 2010; 51:60–70.

42.

Doney

, Cannon

. An examination of the nature of trust in buyer-seller relationships. Journal of Marketing, 1997; 61:35–51.

43.

Chin

, Newsted

. 1999. Structural equation modeling analysis with small samples using partial least squares. Hoyle

. Statistical strategies for small samples research. Thousand Oaks, CA: Sage Publications, 307–341.

44.

Henseler

, Ringle

, Sinkovics

. The use of partial least squares path modeling in international marketing. Advances in International Marketing, 2009; 20:277–319.

45.

MacKenzie

, Podsakoff

. The problem of measurement model misspecification in behavioral and organizational research and some recommended solutions. Journal of Applied Psychology, 2005; 90:710–730.

46.

Ringle

, Wende

, Will

. 2005. SmartPLS 2.0 (beta), computer software. University of Hamburg www.smartpls.de. 2013 May 8.

47.

Keil

, Tan

, Wei

et al. A cross-cultural study on escalation of commitment behavior in software projects. Management Information Systems Quarterly, 2000; 24:299–325.

The Persuasion Context and Results in Online Opinion Seeking: Effects of Message and Source—The Moderating Role of Network Managers

Abstract

Abstract

Introduction

The persuasion process in the search for online opinions

The moderating role of the kind of online opinion network

Materials and Methods

Information gathering and characteristics of the sample

Measurement of variables

Results

Discussion

Management implications

Limitations and further research

Acknowledgment

Notes

Footnotes

Author Disclosure Statement

References