3.1. Data and data coding
As a first step, each exhibit in nexus was coded with regard to its
position in the gallery, its visible features, and its prominence, as
can be seen in Table 2. Due to exhibit loans to other special exhibitions
of the LiMo and due to exhibit exchanges, some data was
missing. Hence 1208 of 1310 exhibits were considered in the
following analyses.
An exhibit's position in the exhibition was coded by two parameters:
showcase number and vertical position. Showcase
number represents an exhibit's horizontal position along the way
through the gallery. Similar to the analysis reported in Study 1, all
39 showcases were numbered from the entrance to the exit along a
linear path through the exhibition, with higher numbers indicating
later placement along the path. Vertical position represents the
position of an exhibit on eye level versus above/below eye level. The
glass shelves within each showcase were numbered. Each showcase
consisted of five shelves. For these shelves, we coded whether
they were on eye-level for average sized persons. We assumed that
Shelves 1 and 5 were not on eye-level because they were either just
above the ground (i.e., Shelf 1) or above an average sized person's
eye-level (i.e., Shelf 5). Consequently, it was assumed that Shelves 2,
3, and 4 were on eye-level. For the regression analysis, we coded
the shelves that were not on eye-level with 0 (N ¼ 485) and the
shelves on eye-level with 1 (N ¼ 743).