Page 141 - JOURNAL OF LIBRARY SCIENCE IN CHINA 2015 Vol. 41

140 Journal of Library Science in China, Vol. 7, 2015

which surprise has the highest correlation. Existing IS researches on user affections mainly
analyze the relationship between positive affections like happiness (Beaudry & Pinsonneault,
2010; Noteborn et al., 2012), excitement (Beaudry & Pinsonneault, 2010; Jiang et al., 2010) and
interest (Y. B. Lu & H. M. Xu, 2006; Van der Heijden, 2004) and users intention or behavior
of information system, and consider the three affections major affective factors affecting user
behavior in a information system. The research in this paper finds that happiness, excitement and
interest are indeed associated with user behaviors in a information system, yet surprise is the most
important factor affecting user satisfaction and user behaviors.
Different from the hypotheses, the negative correlation between negative affections and user
satisfaction is confirmed. The result of the correlation analysis in Table 3 indicates that there is no
significant negative correlation to between negative affections and user satisfaction. The result of
regression analysis in Table 4 indicates that the regression coefficient of negative affections and
user satisfaction is-0.039; it is not significant at the 0.05 level, indicating that there is no significant
negative effect relationship between negative affections and user satisfaction. With detailed
analysis of the relationship between each specific affective state and user satisfaction, we find that
all negative affections, including restlessness, boredom, frustration, anger, sadness, anxiety and
puzzlement, have no significant negative correlation with user satisfaction.
At present, in the studies of consumption affections, the relationship between negative affections
and consumer’s post-purchase behavior remains debated. Some researchers believe that there
is a negative correlation between negative affections and consumer’s post-purchase behavior,
while some researchers hold the opposite view (Geng & S. Q. Chen, 2006). Oliver (1993) divides
negative affections into affections evoked by others, affections evoked by the environment and
affections evoked by the user him/herself. Dubé and Menon (2000) believe that only when evoked
by others (service providers) will negative affections lead to a negative impact on post-purchase
behavior. Through further interviews, we find that micro-blog posts that can produce negative
affections are the main reason causing users to experience negative affections, which is not related
to the micro-blogging platform, namely supportive environment of micro-blog content. This may
explain the empirical result of Hypothesis 2.
Our study finds that user experience has a regulatory function on the relationship between user’s
affections and satisfaction. As the experience accumulating, the relationship between positive
affections and satisfaction is gradually strengthened (i.e. the regression coefficient increases).
This indicates that, compared with new users, old users are more concerned about the affective
experience when using micro-blog, especially positive affective experience. For older users, not
much positive affective experience can bring greater improvement to user satisfaction.

7 Contribution of the research and follow-up work

By studying the relationship between user’s affections and satisfaction, we cover the shortage that