ELM The Elaboration Likelihood Model (ELM) developed by Petty and Cacioppo (1981, 1986) provides a useful framework for understanding the effectiveness of persuasive communication. Based on the ELM, attitude change may occur via two routes of influence, the central route and the peripheral route. The two alternative routes differ in the amount of thoughtful processing of information or elaboration. Individuals taking the central route think critically about issue-related arguments and scrutinize the merits and relevance of those arguments before forming an attitude about the advertisement or product. Conversely, individuals taking the peripheral route make less cognitive effort and rely on shortcuts such as the number of arguments and physical attractiveness of endorsers when forming an attitude (Petty & Cacioppo, 1986). Petty and Wegener (1999) suggest that individuals in high elaboration likelihood state are more likely to engage in thoughtful processing of information and are also more likely to be persuaded by argument quality. People in low elaboration state tend to base their attitude change on peripheral cues. Elaboration likelihood moderates the effects of argument quality and peripheral cues on attitude change. According to the ELM, elaboration likelihood is determined by an individual’s motivation and ability to elaborate. Motivation refers to the individual’s personal relevance to the persuasive message while ability is manifest in the individual’s cognitive competence or prior expertise with the attitude object. Individuals vary in their ability and motivation to elaborate.

Skepticism Skepticism refers to the tendency toward disbelief (Obermiller & Spangenberg, 1998). Consumers come to believe or disbelieve marketing messages through socialization and purchasing experiences. Although the Internet provides easy access to information about products and services, it may foster consumer skepticism. Ford, Smith, and Swasy (1990) reviewed issues about the dimensionality of skepticism. Their study shows that one may have different levels of skepticism toward claims made in advertisements and the motives of advertisers. In other words, skeptical tendency may vary depending on product type or audience personality. Consumer skepticism reflects a marketplace belief held by a consumer. This tendency is a stable, generalizable belief which reflects a consumer’s implicit views of how the marketplace works. A scale measuring consumer skepticism was developed by Obermiller and Spangenberg.

Friestad and Wright (1994) developed a framework for examining the moderating influence of consumer persuasion knowledge in responses to marketing communications. Persuasion knowledge was defined as the learning that consumers use to interpret, evaluate and respond to persuasive attempts from advertisers and marketers. It may include knowledge about persuasion techniques and marketer tactics. Advanced persuasive knowledge implies higher levels of skepticism toward advertising claims. Obermiller and Spangenberg (1998) argue that highly skeptical consumers may perceive smaller differences in situational factors (such as quality and quantity of messages) than do less skeptical consumers. Highly skeptical consumers may be more prone to disbelieve certain categories of product information (such as advertisements or online reviews) without considering the source credibility, argument quality or number of arguments. In other words, highly skeptical consumers may be impossible to persuade via argument quality as they would not believe any stated claims. Unlike consumers with high need for cognition, a highly skeptical consumer shows a tendency to stereotype, where “… stereotype is . . . [a] set of beliefs about the characteristics or attributes of a group” (Judd & Park, 1993). Stereotypes do not need to be inaccurate or negative. In other words, the highly skeptical consumer takes intrinsic shortcuts, which are based on beliefs, to evaluate information instead of using extrinsic shortcuts, which are based on situational factors such as argument quality, to form attitudes. From the ELM perspective, this response to information implies that one fundamental route to persuasion is closed. Therefore, the following hypothesis is proposed:
H1: The purchasing intention of consumers with high skepticism is not influenced by argument quality and quantity of online reviews.

Consumers with low skepticism are characterized by incomplete persuasion knowledge. Friestad and Wright (1994) further characterized consumers with incomplete persuasion knowledge as naïve and less mature, a cognitive status indicative of a less developed cognitive ability. Based on the ELM theory, elaboration likelihood is also determined by the cognitive ability of consumers. Therefore, low skepticism consumers are more prone to be persuaded by peripheral than central cues. Thus, less skeptical consumers who have less persuasion knowledge will behave in ways similar to people low in need for cognition. Individuals with low need for cognition do not enjoy cognitive efforts and prefer to rely on opinions of others (preferable experts) when dealing with complicated issues. They also tend to base their attitudes on simple cues such as the source attraction (Petty, Cacioppo, & Goldman, 1981) and the number of arguments offered by the messages (Petty & Cacioppo, 1984).
H2: The purchasing intention of consumers with low skepticism is more influenced by argument quantity than quality of online reviews.

Subjects were instructed to respond to measures of manipulation checks before taking dependent measures – purchasing intention – and demographic questions. The dependent measure of subjects’ purchasing intention (PI) contained two questions: (1) How likely are you to choose “A Model” cell phone next time you plan to purchase a cell phone of similar nature?; (2) Would you recommend “A Model” cell phone to your friend? These two items were also anchored on a 7-point Likert scale. Finally, subjects were asked to rate the 9-item consumer skepticism measures developed by Obermiller and Spangenberg (1998) on a scale of 1-7 (1 – disagree strongly and 7 – agree strongly) before responding to demographic questions.

Contradicting the ELM theory, the effect of argument quality x skepticism interaction was not significant. The result suggested that high skepticism subjects do not take the central route in formulating purchasing intention. On the other hand, a significant quantity x skepticism interaction emerged from the three-factor analysis [F(1, 270) = 5.662, p < 0.05]. Specifically, the main effect of review quantity manipulation was qualified by a quantity x skepticism interaction. This finding was consistent with the ELM prediction. In other words, the quantity x skepticism interaction indicated that the subjects with low skepticism take the peripheral route in formulating purchasing intention. Simple main effects tests were conducted to further explore the effect. The results revealed that low skepticism subjects expressed more positive attitudes after exposure to the large quantity version (M = 4.83) than after exposure to the small quantity version (M = 4.02) of online reviews [F(1, 127) = 17.755, p < 0.001], but that attitudes of high skepticism subjects exposed to large (M = 3.69) and small (M = 3.54) review quantity did not differ.

In summary, the purchasing intention of online shoppers with high skepticism was not influenced by argument quality of online reviews. In addition, the purchasing intention of online consumers with low skepticism was more influenced by argument quantity than quality of online reviews. Thus, hypotheses 1 and 2 were supported.

The purposes of the present study were to explore the effects of positive online reviews on the attitudes of online consumers using skepticism as a moderator, and the mechanisms supporting the effects. An online experimental study was designed. Two major findings emerged from results. First, in our study people with high skepticism shopping online did not take the central route in attitude change as predicted by the ELM. Highly skeptical consumers tended to base their attitudes on intrinsic beliefs instead of extrinsic factors; that is, they were biased against a certain type of information and indifferent to the message quality. Consumers with high skepticism had a tendency to disbelieve certain categories of product information without considering the source credibility, argument quality or number of arguments. In other words, highly skeptical consumers may be impossible to persuade via argument quality as they would not believe any stated claims contained in online reviews. This finding contributes to the ELM research literature by considering a potentially important personality factor in the ELM framework. Marketers and software developers should identify the profiles of skeptical consumers carefully in order to communicate marketing messages more efficiently. Demographic profiles about people with various personality attributes (such as high and low skepticism) are important for market segmentation and targeting.

Second, low skepticism consumers tend to adopt the peripheral route in forming attitudes in an online context; that is, they are more persuaded by quantity of online reviews since more reviews are perceived as indicating greater product popularity. In this sense, low skepticism consumers have similar characteristics to those with low need for cognition; that is, less skeptical consumers are prone to be persuaded by peripheral cues such as source credibility and number of arguments. This finding contributes to a better understanding of the mechanism underpinning the effect of online reviews. If they wish to influence these consumers more effectively, marketing executives and software developers should articulate a more user-friendly interface that simplifies the process of posting reviews and therefore induces the participating intention. Incentives such as prize draws and interactive online games can also enhance online shoppers’ intention to post reviews. Moreover, knowledge of the consumers’ level of skepticism can guide the internet marketers and software developers alike to design promotional materials and review formats that will influence online shoppers effectively.