International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B6. Istanbul 2004
teachings on the basis of their own naive concepts, conquering
the whole field in small cognitive steps to achieve at the end a
more or less complete scientific understanding.
The MLBK is an interactive didactic program for method
teachings and statistics, which tries, with the help of didactical
interventions, overcoming the phenomena “fear of statistics”
particularly under scholars in humanities and social sciences.
In the core of the software system interactive exercises are
located for self learning, which are developed purposeful to
address and diminish the cognitive and effective problems of
the students when learning method teachings and statistics.
Besides the interactive sessions a book as accompanying text as
well as a glossary exists for looking up terms and improving the
knowledge.
The MLBK supports a computer-assisted learning process,
where the scholars decide on the speed of the learning progress
and the place for learning. Thus, the project MLBK is an
important contribution to the reform of study and teachings,
because:
- the method teachings including statistics is a training item in
many disciplines,
- the modules of the program cover the main topics in various
courses of studies,
- it is possible to structure the sessions in a modular way related
to the specialization area or the complexity degree,
- it is open for internationalisation and extension, and
- representatives from several subjects and several universities
are involved in the project.
The modularity makes a multilingual employment possible in
various specialisation areas. At the same time the MLBK may
be extended and upgraded for future interests and may track the
constantly growing teaching requirements in the different
disciplines easily.
1.2 Overview of the contents
The MLBK is constructed by several components, which
thematically include the majority of the method teaching
context on an undergraduate level in various disciplines. Main
components are the modules from the reality to the data, data
collection, descriptive statistics and inference statistics.
In one special module components for special methods are
integrated such as cluster or time-series analyses. The spatial
visualisation part described here is also part of this special
module. ;
The empirical hands-on training is a part of the learning
environment, where the scholars can deepen their knowledge
gained so far applying it by planning and executing own
investigations (see Figure 1).
Within these modules there are again several layers in the form
of texts, media, exercises and examples, which can be
variegated within the specific content and conditions of the
session.
In the centre of the MLBK interactive exercises for self learning
are located based on the concept of explorative learning. The
idea of explorative learning results from a cognitive analysis of
the learning processes of scholars. This method is suitable to
diminish learning difficulties in complex areas.
Exercises are linked with hypertext-textbooks and hypertext-
glossaries. A central navigation location permits the entrance to
contents from any component of the system. The users are
supported by a monitoring, which on demand informs the user
on completed course parts or positions the user, when re-
entering the teaching program, at the termination point of the
last session.
164
wn
e t L £e
E Y 3 zs
© S 2 = = ih
2 ; e
g EZ = = = = on
= m n S © dc
Io = 2 « u = = =
— os o = e
= ov Zz 9 — € cm
D Sg = = 2 9g
2 8
= =” = > © =
= = 9 qu m e
e Ca 6 c N f
| Texts and examples |
[ Interactive exercises j
Medicine, Education science, Humanities, Psychology,
Sociology, Economical Science
Figure 1: Main structure of the MLBK
2. THE LEARNING MODULE “SPATIAL
VISUALISATION OF STATISTIC DATA"
o
2.1 Technical background
The interactive learning modules for the MLBK are embedded
in an on-line Internet environment. The structure is modular and
allows to construct own learning paths through the different
modules related to the scholars disciplinary background.
Examples and exercises are dedicated to the special disciplines,
but may be selected freely. The framework and the data are
arranged in such a way that they are exchangeable by the
respective lecturer. The high degree of interactivity in this
module can not be developed with simple Internet technologies.
The problem lies on the one hand in the visualisation of the
changeable spatial data with its attributes and on the other hand
with the real-time computations of statistic procedures. Both
requirements could easily be attained with existing software
products, but would cause additional license costs for all
education sites using these modules. Thus, one goal of the
project was to use only cost-free products or self-developments.
This guarantees that the total package can freely be distributed
without auxiliary costs.
Learning contents are stored with the help of an author system
into a MySQL database, which likewise supports the modular
structure. The complex structures are represented on a Website
with the server side script language PHP. Most of the
interactive exercises had to be programmed by our own.
The available subproject for spatial visualisation is developed
based on a specialised example of population statistics. In the
first phase of the project different new technologies were tested
to realise the interactive exercises. We tested clients Java
applications, SVG and Flash elements. Java applications can be
developed with appropriate programming and offer high-grade
interactivity on the client side. For the representation of spatial
data in the web the JAVA-library Geotools is very prominent.
Geotools is a project of the University of Leeds, started in 1996,
and contains an extensive class library for the selection and
representation of spatial and other data. The functionalities of
Geotools could easily be used for the interactive environment.
In the module a set of learning units were implemented with
Geotools. The conversion was quite simple, because only the
different components in a Java applet must be rearranged.
Additional interactions wère developed with the Java library
Swing. All applets are parameterised and are called
dynamically, so that reusability is given. A major performance
problem comes with larger application data sets resulting in
"SR Lt Mea
er