Page 131 - JOURNAL OF LIBRARY SCIENCE IN CHINA 2015 Vol. 41

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

4 Phase one: Affective experience and analysis of the difference

4.1 Methodology and sample

Currently, the main measurement methods for affections are: using scientific instruments to
measure physiological changes related to affections, measuring explicit behavior, as well as self-
report affective measurement (Yan, 2008). The method of measuring with scientific instruments
is objective and precise, but it has higher costs, and ignores the psychological level of affective
experience. The method of measuring explicit behavior is more difficult, because even with similar
affective reactions that are evoked by the same stimulation, each person’s external behavior is
not necessarily the same. The method of self-report affective measurement is based on subjective
reports of the participants that state their affective experience, which is a commonly used affection
measuring method in psychological research.
The traditional advantages/disadvantage method, word association and open task are the three
methods of self-report affective measurement. Dean et al. (2006) compare the three methods, and
find that open task method is the most effective in evoking the participants’ affective response.
This method also has a cross-cultural nature. Through open-ended questions, Page and Mapstone
(2010) ask participants to use their own language to describe their affective reaction to the website
of Youtube, and demonstrate the effectiveness of this method.
In the study of affections, the questionnaire method is more effective than the experimental
method, for it can help people to accurately recall past affective experience, which has been
verified in the study of affections in consumption (Geng, 2007). Therefore, in the first phase we
used a self-report questionnaire measuring method. It is based on a wide range survey of micro-
blog user’s affections with reference to Page and Mapstone’s (2010) affection measuring method.
It asks participants to recall and fill in the most impressive affections in last month’s use of micro-
blog, and measures micro-blog user’s satisfaction at the same time. User experience and basic
demographic information will also be investigated.
We created a questionnaire on www.sojump.com, and generated a link of the questionnaire. We
then posted the link in micro-blogging platforms like Sina and Tencent, inviting micro-blog users
to fill out the questionnaire. Within three months (from December 3, 2013 to March 2, 2014), we
collected 423 questionnaires. By excluding 73 invalid questionnaires, we finally obtained 350
valid questionnaires. We used the independent sample T test method in the mean value analysis
to test whether the samples are from the same population. The T test process is to compare the
mean values of indicators of two samples; if the mean values have significant differences, the two
samples are from different population. 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.