3
In the present study, we attempt to address the above issues. Specifically, we aim to find
out how perception of a source affects individuals’ attitude toward online health information by
the source, health behavior intentions, and attitude toward the message channel. We examined
the differential effects of a commercial source vs. a government (non-commercial) source
through health information on a Web site. Secondly, we aim to examine if a characteristic of an
information channel moderates the effects of source. We are particularly interested in if and how
sources of online information benefit from interactivity of the Internet. Finally, we aim to
uncover a mechanism in which the channel characteristic (i.e., interactivity) influences the
effects of source.
SOURCE EFFECTS
Since the seminal work by Hovland, Janis and Kelly (1953), Kelman (1961), and
McGuire (1985), the role of an information source in persuasion has been widely recognized and
the large number of source effects was documented in the literature across disciplines (Artz,
1995). Scholars in social psychology, communications, and consumer behavior have been
especially keen to a wide range of outcomes of source and source characteristics because
individuals’ perception of a message source is a major determinant of whether individuals accept
or resist a persuasive message (Andrews & Shimp, 1990; DeShields, Kara, & Kaynak, 1996;
Wilson & Sherrell, 1993).
According to McGuire (1989), “Source variables in persuasive communications refer to
characteristics of the perceived communicator to whom the message is attributed, and not of the
person……who actually produced the message” (p.46). However, other researchers provide a
more broad definition of source, which includes “the originator of the message, the channel
through which the message is delivered, and an endorser who is featured in the message”
(Barnes, 1978, p.235; Fuchs, 1964). Acknowledging various components of a source, these
researchers argued that source is “n-dimensional” and that a number of sources, or “total source,”
are working simultaneously to influence individuals (Barnes, 1978; Fuchs, 1964). How the total
Convention