Cross-Cultural Design and Evaluation Framework - Cross-Cultural Human-Computer Interaction and User Experience Design - page 86

7.2 HYPOTHESES AND QUESTIONNAIRE
For the pilot study, we gathered 45 hypotheses about the specifics of Chinese users. The
hypotheses were drawn from the conclusions of previous research in usability testing,
psychological studies, visual semiotics, and linguistics. For the information value of
UI components's spatial organization, we worked with the oppositions of Given/New,
Ideal/Real, and Center/Margin proposed by Kress and van Leeuwen (2006). In this
context Given is taken-for-granted information and New information is introduced
later. Similarly, Ideal presents what might be, and Real what is, for example, a specific
or practical information. All of these oppositions can be combined with Center, a
nucleus of information (or the most important information), and Margin, containing
other dependent information.
Some of the conclusions were directly included in the hypotheses (e.g., favorite
colors for the background: blue, purple, cyan, gray), some were modified according
to our assumptions (e.g., because Given information in the West is expected on the
left of a screen, we expected the information to work better on the right in China),
some were constructed from our direct experience with the Chinese culture (e.g.,
red color with yellow text is used for special occasions), while some tested our
more general assumptions (e.g., there is a close similarity between the sequential
information structure in language and the horizontal structure in visual composition).
Some of the hypotheses that came from the above process of construction did
not perform well in the early tests, or were too vague to translate into a question,
that would be meaningful in the context of the study. Such hypotheses and the related
questions were omitted from the research. The complete list of hypotheses is available
in Appendix C, “Hypotheses.”
In order to test the different hypotheses, we created at least one question for each
hypothesis. The questions were grouped according their areas, that is, personal infor-
mation, layout, color, symbol, and look and feel. Where appropriate, we used closed
questions. However, because we carried an exploratory pilot study, we did not want
to constrain our respondents and offered mostly open questions. The questionnaire
was supported by examples of UI components.