EGIS (1994), copyright EGIS Foundation.
The national institutional conditions for the diffusion of GIS are examined and the time lag in the production of digital maps in Germany is explained. The paper indicates the need for appropriate locational referencing standards. To analyse the settings at the adopter level some of the results from a survey of GIS adoption are presented. The findings indicate that the institutional and organizational settings are critical for the degree of GIS utilisation.
It is generally accepted that there are great differences in the diffusion of geographic information technologies in Europe. Whereas some countries have reached a considerable high level of GIS utilisation other countries are slow in adopting the new technology. This paper describes some aspects of the GIS diffusion process in Germany. In the first part, the national settings regarding digital maps and socio-economic data are evaluated. In the second part of the paper a research approach in the field of analysing the utilisation process at the adopter level is described.
Britain's Chorley Committee on Handling Geographic Information (DoE 1987) mentioned a number of factors which inhibited the take-up of GIS in Britain. Besides the availability of technology the most important findings are related to digital topographic mapping, the availability and linking of data and the role of the government. As a result of the report the Ordnance Survey (as the only responsible agency for mapping) modified its digital mapping specification in order to reduce costs and hence complete the digital conversion sooner (DoE 1988). Consequently at the large scale all of the 1:1250 scale, 60% of the 1:2500 scale and some of the 1:10000 scale (full cover by 1995) maps were available in digital form in November 1992 (OS 1993).
In Germany there is a much more complicated situation. In order to understand why it is essential to describe the institutional organisation of mapping. Like in other European countries there exists not only one surveying agency. The responsibility for mapping lies at different levels. Firstly, local government
[End Page 1503]
surveying departments are engaged with 1:1000 city maps and with the maintenance of the cadastre. Secondly, the state surveying agencies (Landesvermessungsamter) of the 16 Bundeslander are responsible for the medium-scale maps (range from the 1:5000 to the 1:100000 scales) of their areas. Finally, the Institute of Applied Geodesy (Institut fur angewandte Geodasie) in Frankfurt is in charge with the small-scale maps (1:200000 or above).
The German land information system traditionally consists of two parts: the cadastre (Liegenschaftskataster) which offers information on location and size and the Land Register (Grundbuch) which contains the property of the land parcel. Already in the 1970's, in order to automate the maintenance of both registers, two concepts were launched: the Automated Cadastral Register (Automatisiertes Liegenschaftsbuch, ALB) and the Automated Cadastral Map (Automatisierte Liegenschaftskarte, ALB). Until the end of the 1980s only little progress has been made in implementing the ALK system.
From 1985 to 1989 the state surveying agencies have developed the concept of the medium-scale map project ATKIS (Amtliches Topographisch-Kartographisches Informationssystem ). The ATKIS data model is based on the object specifications developed for the ALK data base (Christoffers 1992). Two different states are projected to store the topographic data: digital landscape models (Digitale Landschaftsmodelle, DLM) which contain the objects of the landscape and digital cartographic models (Digitale Kartographische Modelle, DKM) which hold the cartographic signatures. In a first step, the DLM 25/1 is build up. It contains 65 of the originally 175 object codes.
The state surveying agencies have begun the realisation of the DLM 25/1 at different points of time and with different priorities. Only 4 of the states, Nordrhein-Westfalen, Niedersachsen, Rheinland-Pfalz and Bayern (GEOGIS) started before 1991 (Christoffers 1992). The national coverage is not available before 1996.
To meet the increasing demand for digital maps, the surveying agencies have also started scanning and digitising some of their paper maps. In most cases this took place not earlier as 1990. In summary, there are striking differences between the German Bundeslander regarding the extent to which digital maps are available.
A second national dimension that has to be taken into account is the availability of geographic referenced digital statistical data. The GIS revolution stimulates the general discussion about the aggregation form of the data supplied by the statistical and other governmental agencies again. Due to data protection reasons information is only available on an aggregated level which makes the use for spatial analysis difficult or impossible. Until now, very little progress has been made towards a national referencing system. Today, the administrative boundaries are the most used system for socio-economic data.
[End Page 1504]
The third national characteristic considered is the situation of education and training in GIS (DoE 1987). The study of Bill (1992) emphasises the unsatisfactory situation of GIS education in Germany in 1990. During the last two years some progress has been made in establishing GIS at universities (Institute fur Photogrammetrie 1992).
Before arriving at the decision whether to adopt an innovation or not the potential adopter passes through an information seeking process in order to maximise his benefit. Due to the difficulties in estimating the benefits of information systems earlier adopters reduce the risk of the innovation for later adopters. That is the basis for the high importance of interpersonal channels at the persuasion stage in the innovation-decision process. According to that, to estimate the diffusion process of an innovation in a social system on an early stage it should be an appropriate approach to analyse the utilisation process of all adopters.
There are three main groups of problems: technical, data-related and organisational difficulties.
The role of human and organisational issues in the process of GIS utilisation is generally recognised (DoE 1987; Openshaw et. al. 1990; Campbell 1991). In their current state of development most of the commercial GIS packages are universal technologies, representing tools than can be applied in a variety of ways to diverse problems. Geographic Information Systems as the combination of hardware, software (i.e. the commercial GIS as the core and a set of analytical or statistical packages), data and organisational settings are modified or changed by an organisation during the process of adoption and implementation in a way that fits its conditions best. The organisation goes through a period of problem-solving after initial purchase. According to that, the research on the diffusion has to study the whole implementation process, not just the adoption decision (Rogers 1993).
Campbell (1991) summarises the organisational conditions for the effective implementation and utilisation of computer-based systems authorities as the existence of an overall information management strategy, the personal commitment and participation of individuals and a high degree of organisational stability.
The research has been concentrated on public sector organisations, especially on local government as one of the most important groups of users of GIS (Campbell and Masser 1992). In certain things (i.e. size and functions) the situation of local government in Germany differs from other European countries. Therefore and due to the overall slow rate of adoption the unit of analysis in this study are all acquirers of a commercial GIS.
[End Page 1505]
82 acquirers of a GIS were mailed using the client list of the system vendor. A total of 42 questionnaires were returned (response rate of 51%). Some of the respondents are users of a mapping software of the system vendor, they were excluded from the further analysis. Table 1 shows the type of organisation for the remaining 38 respondents. It can be seen that nearly 40 percent of all GIS adopters are private firms.
Table 1 GIS Adopters by Type of Organisation
Number %
Authority 10 26.3
University 10 26.3
Private Sector 15 39.5
Association 3 7.9
Total 38 100
The respondents were asked for which application areas they use their GIS. Multiple answers were possible. Table 2 analyses the findings. For all of the non-university users the main application lies in the field of Landscape Planning and Environment Monitoring.
Table 2 Application Areas by Organisations
Authority University Private Assoc. Total
(%) (%) Sector(%) (%) (%)
Regional Planning 16.7 9.1 7.3
Town Planning 16.7 5.9 4.5 7.3
Traffic Planning 8.3 9.1 5.5
Landscape Plan. 25.0 5.9 40.9 75.0 29.1
and Environment
Monitoring
Statistics and 8.3 13.6 7.3
Marketing
Education 47.1 14.5
Research 8.3 35.3 4.5 14.5
Other 16.7 5.9 18.2 25.0 14.5
(n=12) (n=17) (n=22) (n=4) (n=55)
(Column Percentages, Multiple answers possible)
The questionnaire motivated participants to think about the acquisition, implementation and use of the GIS within their organization. Participants were presented a set of possible problems that could be related to the implementation of the system. They were asked to what degree each of the problems hindered the implementation process.
[End Page 1506]
Table 3 Average responses on problems associated with the
implementation of GIS (n=38)
1= no problem
3= neutral
5= great problem
Problem Average of Std.
Responses Dev.
1 Availability of digital maps 3.15 1.33
2 Availability of statistical 2.49 1.23
data
3 Compatibility of GIS with 2.91 1.27
existing data sets
4 Vender support 2.44 1.01
5 Manual of GIS 2.71 1.09
6 Ease of learning GIS 2.00 1.25
7 Compatibility of GIS with 2.70 1.18
existing software
8 Hardware relaibility 1.79 0.96
9 Skilled staff 2.71 1.31
10 Managerial structure for GIS 2.06 1.22
implementation
11 Encouragement from senior 1.77 1.18
staff
12 Staff resistance 1.53 1.02
The findings presented in Table 3 provide some insights into the implementation problems experienced by the acquirers. The important issues of digital maps can be seen. Central task of the next step was the classification of the GIS use. Respondents were asked a series of questions on their use of different GIS functions. The algorithm used for determining cluster membership in the procedure is based on nearest centroid sorting. The final cluster centres and the number of cases in each cluster are given in Table 4. The two types of users emerged from the cluster analysis were designated as mapping user (cluster 1) and analytic users (cluster 2).
Table 4 Final Cluster Centres for the 2-cluster-solution
Cluster 1 Cluster 2
Basis functions 4.42 4.67
Digitise 3.58 4.07
Thematic Mapping 3.36 3.87
Distance Operations 1.53 3.60
Polygon Overlay 1.89 4.13
Aggregation 2.11 3.80
Number of Cases in 20 16
each Cluster
In the next step, the average responses on problems by the two user groups were compared. The findings presented in Table 8 indicate a number of important differences between the problems associated with the GIS implementation. Organisational difficulties were most pronounced from the mapping users. The lack of skilled staff stands on the second rank commonly with the software manual (only available in English language). On the other side the greater use of the data analysing functions leads to more problems in linking different data sets.
Table 5 Average responses on problems by the user groups
Problem Mapping Analytic t
User User scores
1 Availability of digital maps 3.2 3.0 0.4
2 Availability of statistical 2.4 2.5 -0.1
data
3 Compatibility of GIS with 2.5 3.4 -2.1*
existing data sets
4 Vender support 2.6 2.1 1.5
5 Manual of GIS 2.9 2.5 0.9
6 Ease of learning GIS 2.2 1.6 1.3
7 Compatibility of GIS with 2.4 2.9 1.2
existing software
8 Hardware relaibility 1.9 1.5 1.4
9 Skilled staff 2.9 2.2 1.6
10 Managerial structure for GIS 2.3 1.4 2.3*
implementation
11 Encouragement from senior 1.9 1.5 0.8
staff
12 Staff resistance 1.6 1.1 2.4
(n=20) (n=13)
(* t scores significant to the .05 level)
The part of the empirical findings presented in this paper provides only a first insight in the implementation and utilisation process after acquisition of the GIS. Including the time dimension (i.e. year of purchase) shows that the analytic users have much more experience in handling GIS.
Bill, R. (1992), on the Situation of GIS-Education at German Universities, Proceedings of Third European Conference on Geographic Information Systems, EGIS '92 Munich, Volume 2, pp. 846-854.
Campbell, H. (1991), Organizational issues in managing geographic information. In: I. Masser and M. Blakemore (ed. ), Handling geographical information: methodology and potential applications, pp. 259-282. Longman: Harlow.
[End Page 1508]
Campbell, H. and Masser, I. (1992), GIS in Local Government: an Overview of Take-up and Implementation, TRP 111 Department of Town and Regional Planning, University of Sheffield: Sheffield.
Christoffers, F. (1992), Rahmenbedingungen zur Einrichtung des ATKIS-DLM 25/1 in Niedersachsen, Nachrichten der Niedersachsischen Vermessungs- und Katasterverwaltung, 3, pp. 121-133.
Department of the Environment (DOE) (1987), Handling Geographic Information, HMSO: London.
Department of the Environment (DOE) (1988), Handling Geographic Information. The Government's response to the report of the Committee of Inquiry, HMSO: London.
Institut fur Photogrammetrie (1992), Vortrage zum Workshop Geoinformationssysteme in der Ausbildung, Schriftenreihe Heft 16 Institut fur Photogrammetrie: Stuttgart.
Openshaw, S., Cross, M., Charlton, M., Brundson, C. and Lillie, J. (1990), Lessons learnt from a Post Mortem of a failed GIS, Proceedings of Association for Geographic Information Conference, AGI '90 Brighton, pp. 2.3.1-2.3.5.
Ordnance Survey (OS) (1993), Digital Map Data catalogue 1993, OS: Southampton.
Rogers, E. M. (1983), Diffusion of Innovations. The Free Press: New York.
[End Page 1509]