Thus, we investigate word-of mouth activity among members of a Friend

Network that leads them to become members of a separate Product Network[2]

on a social networking site. Examination of such activity is important, because research has shown that the types of user behavior that occur among network members once they have joined a product page can lead directly to

product purchase (Trusov et al., 2010).

In a later section, we explore how characteristics unique to online social networks may moderate the effects of the antecedents of WOM influence (e.g. source tie-strength) on consumers’ reactions to a persuasive, product-related communication. First, however, in order to frame online WOM effects in a communications context, we turn our discussion to communication hierarchy response models.

HOEs involving WOM communications in online social networks

A number of response hierarchy models have been developed to depict the stages a consumer may pass through in moving from a state of not being aware of a company, product, or brand to actual purchase behavior. Perhaps, the best known of these response hierarchies is the model developed by Lavidge and Steiner (1961) as a paradigm for setting and measuring advertising objectives. According to this model, recipients of a persuasive communication pass through the following stages:

(1) awareness;

(2) knowledge;

(3) liking;

(4) preference;

(5) conviction; and

(6) purchase.

Awareness is achieved when a site member receives a communication regarding the existence of Product X. That communication may come from:

.someone within the user’s Friend Network;

. someone outside the user’s Friend Network; or

. in the form of an advertisement.

We concentrate upon the first message source in this study, and hereafter refer to the communication itself as a “directive message”. If that directive message (DM) is ignored, knowledge transference does not occur. It is only when the communication recipient is persuaded to view the product page content that knowledge-creating information is conveyed (Wen et al., 2009).

Continuing along the HOEs, in traditional WOM whether a consumer has reached the “liking” stage can only be inferred based on subsequent product adoption, or through some form of intervening assessment. Similarly in e-WOM, liking might be determined using a survey or questionnaire, or by observing pass-along (i.e. e-mail forwarding) behavior. However, there is also precedent for determining product liking based upon click data (De Bruyn and Lilien, 2008)[5]. As noted earlier, what distinguishes product liking in online social networks from that observed in other types of online forums is that in online social networks a user publicly indicates his/her liking of a product by joining the Product Network. Further, as is the case on Facebook, the act of joining a Product Network (i.e. also referred to as “confirming” a product page “suggestion”) may be actively promoted to other Friend Network members via that user’s “News-Feed”.

In terms of the last three stages, preference and conviction may be inferred from the degree of activity and influence that is demonstrated by membership in the Product Network (Trusov et al., 2010). That activity (particularly if one has reached the conviction stage) typically includes pass-along behavior. Finally, the fact that purchase has occurred may be indicated by such activities as posting ratings, comments, or even product photos[6]. As noted earlier, because previous studies have examined the type of user behavior that occurs among that sub-group of social network site members that have already developed an interest in a particular product page (i.e. Product Network members), we concentrate on the WOM activity that drives them to the page in the first place. In other words we target our discussion toward those factors that drive awareness, knowledge, and liking in an online social networking environment (Figure 1).