International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B6. Istanbul 2004
Sted
-3Sp. Methudec
flGsau s Sov
Leder
A Konzenteitian-
Race Visushtiorurs
Eng
Race und Aus
3 Ef. Kantogt: Casto
22 uf Qe antsprechonden Karte dar
Übueq: veu
a ET DES
AL Sm s E
Figure 2: Visualisation of absolute and relative values
A meaningful choice of the colours is particularly important
with thematic evaluations. Colour-optically and psychological-
correctly selected colours facilitate the legibility and
interpretation ability of the map. Depending upon application
sliding colour scales or opposite colours should be used. Sliding
colours are suitable for values of equal sign. Opposite colours
express positive and negative effects better. Alternatively they
can be worked out with hatching. Opposite values can be
represented also very well with traffic light colours. Apart from
this well-known red-yellow-green colour combination one uses
frequently blue-white-red, because this corresponds to the
human sense for cold and warm. Similar to the first lessons the
scholars work on the same question again, but now they are
applying spatial visualisation techniques. By this, the student
learns, which representation form is suitable and that
cartographic mapping makes phenomena and relations such as
population movements (like rural exodus or suburbanization)
visible.
The formation of classes becomes necessary if the entities,
being either metrically scaled or with constant or discrete
characteristics, show a wide variety of possible values. The
class formation allows to recognize the characteristics of the
distribution and to determine regularities in the data sets. In
such a lesson the student learns, how many classes of
meaningful values may be constructed and how many classes
humans are able to separate visually. The number of classes and
the class borders affect the derived measures as well as the
visualisation.
Municipalities are the smallest units in the official statistics.
They can be hierarchically combined into larger administrative
units by means of aggregation and later be also disaggregated
from larger units to smaller units. In Germany there are
administrative subdivisions, which follow the international
framework of statistics. Each municipality is assigned to an
districts and a country. The key of the municipality is used to
distinguish between the municipalities. Additionally there is an
allocation in regions, which may be governmental districts or
planning regions. Several districts may belong to a region,
again a hierarchical relation. In this lesson the scholars explore
the fact that certain absolute values may be aggregated from
one administrative level to the next level. On the other hand
relative values such as population by area can not be aggregated
directly. Therefore one has to go back to absolute values for
aggregation and calculates the relative values afterwards. The
students also recognize that aggregation and classification
affects the measures as well as the graphical visualisation.
166
D
<< C'asgrégoren,
GJ Seneertall
Gemeinde
“I. ^ amt
51) Zeitreihen Li
CD valuation ur
LE Pe Land
wot Such
CO sten
CI Symbole
Übung: Wählen Sie nacheinander die verschiedenen Aggregationsstufen und beachten Sie den Einfluss
einzelner Einheiten auf die Wertigkeit der nachsten Aggregationsstufe,
4» i uf az i |
AN mese] M jb d ua ü
Figure 3: Evaluation representation aggregation
Population statistics is collected for a long period of time which
allows answering questions on population changes over time. In
the last lesson the scholar is working on time-series. These are
usually provided for one period from one year and made
available by the official statistics. To check longer periods in
time new sets of values such as balance rates are computed. For
the representation of the time series existing procedures known
from the visualisation of absolute and relative measures can be
used. For absolute values these may be diagrams or cartograms
on a map as well as signatures. Relative values may be
indicated in sequences post or next to each other. Alternatively
a relative characteristic may be determined for the respective
period of time. Nearly all presentation methods can be
combined with each other.
"elle t voa
(LI Raümidalen ünd Auge
Et Kartagr. Darateliu
ocinhetere
DAggrégites -
CG Sétiserta
SI Zoitraihen
ka Mekal
oc mal
Mit Hilfe von karto- und Diagrammen können mehrere at]
n dargestelit werder
Übung: Wahlen Sıe mehrere Merkmale und dazu eine pa!
i hg der Veranderung
der Bevolkerung in einem Kreis. i {
i ; j «pg 3 $c € d a
Figure 4: Time-series analyses
By means of diagrams or cartograms several absolute values
and temporal successions can be represented. For the direct
comparison of the regional areas over time the values for
arbitrary periods (e.g. balance rates from 1995 to 2000) can be
computed. Thereby the relative values always refer to the last
conditions of the respective statistical collection. With such
derived relative measures again meaningful visualisations can
be created.
3. FIRST EXPERIENCES
In the context of the standard lecture on geoinformatics in a
course of studies on land management and environmental
protection (32 students in the sixth semester) we evaluated the
learning module “Spatial Visualization of Statistic Data” in
summer 2003. 22 students participated in the evaluation.
Statistics is not a central component of that lecture and the
statistical knowledge of the undergraduates in this engineering
course are not comparable to those of the targeted courses of
