Page 134 - JOURNAL OF LIBRARY SCIENCE IN CHINA 2015 Vol. 41

Luchuan LIU & Kai SUN / Analysis of the relationship between microblogging users’ affections and users’ satisfaction 133

Through Phase one, we found 25 kinds of affections that users have experienced when using
micro-blogs and the differences in the affective experience between different age groups.

5 Phase two: Correlation and regression analysis of affections and satisfaction

5.1 Methodology and samples

By analyzing the data of phase one, we find that distribution of the words of affections from the
most frequent “happy” (41.7%) to the least frequent “stress” (0.6%) has significant discrepant.
Therefore, in the analysis of phase two, we select the first 16 kinds of affections (at least 1%
of participants report they have experienced the affections). By using Westbrook’s affection
measuring method and Likert’s seven-level scale, we measured the frequency of each affection
that each participant experienced in the past week (Westbrook, 1991). The scale is ranging from
“never experienced” to “experienced whenever I use it”. We also measured user satisfaction
with Bhattacherjee’s satisfaction measuring scale (Bhattacherjee et al., 2008). The four items,
“very fulfilled”, “very comfortable”, “very pleasant” and “very satisfied”, are used to measure
user satisfaction. Mean value of the four items is used to suggest user satisfaction. We sent
out questionnaires via www.sojump.com, and finally collected 210 questionnaires. Excluding
questionnaires answered by the same IP address, questionnaires answered in less than 75 seconds
and questionnaires that have the same option for all questions (Y. Liu, Lang, & L. Li, 2013), we
finally collected total 155 qualified questionnaires.
Result of the T test shows that sorted by the time of filling out the questionnaire, the top 20%
the samples and the last 20% of the samples have no significant differences (α=0.01) on all the
measuring items, which means the samples are from the same population.

5.2 Data analysis

As shown in Table 2, positive affections include nine affections of happiness, interest, excitement,
fullness, ease, novelty, surprise, likeness and freedom, and negative affections include restlessness,
boredom, frustration, anger, sadness, anxiety and puzzlement. In positive affections, the mean
value of all affection is greater than 4.8, and the mean value of positive affections is 5.23. Except
for excitement, the mean value of all positive affections is greater than 5. In sharp contrast, the
mean value of all negative affections is less than 3.3, and the mean value of negative affections is
3.1. The mean value of user satisfaction is 3.82 with the maximum value of 5, which means sample
users’ satisfaction is above average.