 |
 |
 |
 |
|
 |
 |
|
 |
Another problem inherent in on-line research is the inability to calculate
traditional response rates, or even estimate non-response, as compared to survey
research through the mail. It is not impossible to assess the exact number of
individuals who have access to a Web site and thus, who represents a sampling
frame for response rate calculations. However, our commercial on-line research
partners have generally been less concerned with response rates than response
quality; they have no trouble recruiting participants for their studies. The primary
issue is obtaining samples that are consistent demographically with their key
audiences.
A related concern is the limited number of respondents in both parts of the
study. Given that just under a 100 respondents participated in Part One and 343 in
Part Two, the findings from this study should be verified with a larger sample to increase
generalization. Another limitation is that the data in both stages of the study were obtained
from respondents at individual Web sites (HotWired in Part 1 and AOL in Part 2).
Although the results do not appear to be biased in any site-specific way, future research on
uses and gratifications for the World Wide Web should attempt to reach users
through a wider range of sites.
Nevertheless, the initial list of 179 potential uses and gratifications presented here,
along with the five underlying dimensions, provide a basis from which to move forward in
future research on uses and gratifications for the Web. For example, future research
can use the current study’s results for classic U&G measure development (e.g.,
Bantz, 1982; Levy & Windahl, 1984) or to segment Web and computer users based
on their individual motivations.
Moreover, this research began the process of attributing certain motivations
to users with a higher affinity for the computer and those individuals who use the
Web more frequently. Results suggest both process and content gratifications are
important, and different motivations relate to affinity with the computer and
frequency of Web and computer use. Results also suggest that process and
content