Geoscience Reference
In-Depth Information
Many analyses of open answer questionnaires use the tools developed by
Muller and Benzécri, for example (Tomasetto et al. 2008). Our analysis of
the users' survey is also based on these tools and is demonstrated through a
process example, the specification of trekking maps. It includes different
steps; first, setting up a users' survey which makes it possible to identify
the specific needs of a community and the user's wishes, then, defining
maps (data and symbolisation) that correspond to the survey results.
2.1 The users' survey
The users' survey is more precisely described in (Baldit-Schneller and
Dominguès 2010). Four groups have been interviewed according to their
activity: walking, road cycling, cross-country biking and finally, those who
occasionally practice these activities. The interview has been conducted by
a coordinator and open-ended questions (the same ones for every group)
have been used. The aim was first, to identify the relevant objects and con-
cepts for trekking maps and secondly to identify the user's criticism and
expectations about maps within his particular activity.
Four French written survey sub-corpora (including, more or less, 46 000
words) have been compiled by using transcripts of the questions and their
replies. Every sub-corpus describes each user's sporting or other activity.
Therefore, it is possible to identify the relevant concepts and the user's
expectations for every activity. The sub-corpora have been studied with
natural language processing (NLP) and lexicometry methods and tools.
2.2 The user's opinion
The lexicometry tools ensure searching the significantly frequent words in
every sub-corpus compared to the reference corpus. In this case, the refer-
ence corpus used is the four sub-corpora set. The statistic comparison
reveals the relevant concepts for every group. The specific words of every
activity are presented in the following table.
