Summary and Conclusion
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
136 137 138 139 140 141 142 143 144 145 146
The results of both studies lend support for the hypothesis that the aggregate method
would produce the most complex collective maps, whereas the congregate method would
produce collective maps lower in complexity. The workshop method tended to produce
collective maps with an intermediate degree of complexity. This result is particularly
evident for the L/N ratio, but is also reflected in several map density comparisons. In terms
of distance ratios between individual maps and the collective maps, the congregate and
workshop methods were less distant than the aggregate method in both cases.
In terms of subjective ratings of the methods, Study 1 suggested that the aggregate
method was rated better than the congregate method for all criteria. Moreover, the
aggregate and the workshop mapping methods were equally good in terms of problem
representation and stakeholder implication. However, the aggregate method outperformed
the workshop mapping method in terms of solution implication. In Study 2 a
different pattern of results emerged. Workshop mapping was generally superior to
aggregate and congregate approaches across all four criteria. Further, congregate maps
generally received somewhat better ratings than aggregate maps. Together the two
studies lend mixed support to Hypothesis 2.
It is useful to consider the relationships between the objective and subjective results.
The objective results were for the most part consistent across the two studies, but there
were differences in the subjective results. Aggregate mapping fared better in Study 1 than
in Study 2, whereas congregate mapping fared better in Study 2 than in Study 1.
Workshop mapping was rated well for the most part in both studies. In Study 1 its ratings
were equivalent to those of aggregate mapping (except for solution implication), whereas
in Study 2 it was rated much higher than aggregate or congregate mapping. This suggests
that map complexity and the degree to which the map corresponds to individual
representations had different meanings for the two samples. In Study 1 map complexity
did not seem to correlate with lower ratings, whereas in Study 2 it did. In Study 1, degree
of difference between collective and individual maps did not correlate with ratings,
whereas in Study 2 it was positively correlated.
At least three explanations for this difference can be advanced. First, and most plausible
to us, the average number of links and nodes was much higher in the individual maps in
Study 2 than in Study 1 (an average of 10 nodes and 15 links in the individual maps in
Study 1 versus an average of 29 nodes and 83 links in the individual maps in Study 2).
Hence, the aggregate maps in Study 2 were likely to be much more complicated and
difficult to interpret than those in Study 1. This may have resulted in lower ratings by the
subjects. In addition subjects in Study 1 may have been able to see their own concepts
and ideas in the aggregate maps more easily than subjects in Study 2, and thus they might
be disposed to rate it higher than in Study 2. Both groups rated the workshop method
high, which suggests that higher levels of participation increase perceived value of the
collective map.
A second explanation is that the use of groups to build maps in all three conditions in
Study 1, but only for the workshop method in study 2, influenced ratings. Participation
in Study 1 may have mediated subject ratings of the maps, particularly those of the
subjects who built the maps. In Study 2, however, subjects only participated in finalizing
the maps for the workshop map and may have seen the aggregate and congregate maps
as “alien.” If this explanation is accurate, then one implication is that the aggregate
method used in a workshop is superior to the other methods although it performs less
well when the facilitator builds the maps. A third possible explanation for the results is
that they stem from cultural differences between the subjects in the two studies.
However, it is not apparent what cultural differences could account for the differences
in results.
Lessons Learned
We can advance several lessons learned about the three methods. The advantage of the
aggregate method is that it is simple and easy to implement. It is also very good at pooling
information from group members’ individual maps. As the comparative study suggested,
the aggregate method works best when individual maps are not very complex and/or when
the group is relatively homogeneous (as in Study 1). The disadvantage of the aggregate
method is that group maps derived from this method tend to be complex and dense. In
larger and more heterogeneous groups, the aggregate method may not be effective as it
is in small and homogeneous groups (as in the setting of this experiment).
It seems likely that the congregate method would be more effective in larger and
heterogeneous groups. The advantages of the congregate method are two-fold. First, it
is less complex (than the aggregate method). An increase in the number of causal loops
does not necessarily increase group map complexity. Second, it is better in representing
the interactions of multiple perspectives via causal loops. In organizational problem
formulation, the congregate method is proposed to apply at the organizational level,
where the number of groups is many and groups have more distinct perspectives. At the
organizational level, the congregate method will help identify different perspectives to
be included in the organizational model.
The workshop method fared best in terms of least complexity and highest subjective
ratings across the two studies. This is likely due, in part, to the higher level of involvement
subjects have in the map-building process. The workshop method can be used alone or
in combination with other methods to improve the shared effect of collective maps. Other
researchers (Ackermann et al., 1997; Diffenbach, 1982; Eden, 1989) have suggested that
such combinations improve the group mapping outcome. The workshop method can
replace the aggregate method at the group level when individual maps cannot be obtained
for some reason. This substitution may not seriously affect the effectiveness of group
maps. In combination with the workshop approach, the congregate method has potential
in handling multiple perspectives in complex systems.
One final point to bear in mind is that Study 1 showed that, in terms of objective measures,
the congregate and workshop methods had more variation in results than the aggregate
method. This is probably due to the fact that these require more judgment on the part of
the modeler and group than the aggregate method, which is for the most part based on
clear-cut rules. Subjective judgment may well be perceived as bias by some participants,
which may create a sense that both the congregate and workshop methods are not
representative of some user perspectives.
Based on the results of the two studies, we can also advance some suggestions
concerning the relationship between group size and effectiveness of mapping methods.
We propose that when group size increases, the effectiveness of the aggregate method
will decrease significantly, while the effectiveness of the workshop method may increase
slightly, and that of the congregate method will increase.
Table 16 summarizes our hypothesized relationship. This relationship may suggest the
best fit method for a given group size. Further study is, however, needed to fully test this
relationship because we are going beyond the four-person groups used in the first study
and the seven-person group in the second.
References
Ackermann, F., Eden, C., & Williams, T. (1997). Modelling for litigation: Mixing qualitative
and quantitative approaches. Interfaces, 27, 48-65.
Table 16. Dependency of method effectiveness on group size
Effectiveness Group size
Rank < 5 5 to 7 > 7 (hypothesized)
Best Aggregate method Workshop method Congregate method
Intermediate Workshop method Workshop method
Worst Congregate
Aggregate method =
Congregate method Aggregate method
Axelrod, R. (Ed.) (1976). Structure of decision: The causal maps of political elites.
Princeton, NJ: Princeton University Press.
Boland, R. J., Tenkasi, R. V., & Teeni, D. (1994). Designing information technology to
support distributed cognition. Organization Science, 5, 456-475.
Bougon, M. G. (1992). Congregate causal maps: A unified dynamic theory of organization
strategy. The Journal of Management Studies, 29, 369-389.
Checkland, P. (1981). Systems thinking, systems practice. New York: John Wiley & Sons.
Courtney, J. F. (2001). Decision making and knowledge management in inquiring organizations:
A new decision-making paradigm for DSS. Decision Support Systems,
31, 17-38.
Diffenbach, J. (1982). Influence diagrams for complex strategic issues. Strategic Management
Journal, 3, 133-146.
Dutta, A. (2001). Business planning for network services: A systems thinking approach.
Information Systems Research, 12, 260-283.
Eden, C. (1989). Using causal mapping for strategic options development and analysis
(SODA). In J. Rosenhead (Ed.), Rational analysis for a problematic world:
Problem structuring methods for complexity, uncertainty, and conflict (pp. 21-
24). Chichester, UK: Wiley.
Eden, C., & Ackermann, F. (1992). The analysis of cause maps. The Journal of Management
Studies, 29, 309-323.
Eden, C., Jones, S., & Sims, D. (1983). Messing about in problems: An informal structured
approach to their identification and management.. New York: Pergamon.
Forrester, J. W. (1969). Urban dynamics. Cambridge, MA: Productivity Press.
Forrester, J. W. (1961). Industrial dynamics. Cambridge, MA: Productivity Press.
Hall, R. I. (1976). A system pathology of an organization: The rise and fall of the old
Saturday Evening Post. Administrative Science Quarterly, 21, 185-211.
Hart, J. A. (1977). Causal maps of three Latin American policy makers. World Politics,
30, 115-140.
Hong, T., & Han, I. (2002). Knowledge-based data mining of news information on the
internet using causal maps and neural networks. Expert Systems with Applications,
23, 1-8.
Klein, J. H., & Cooper, D. F. (1982). Causal maps of decision makers in a complex game.
Journal of the Operational Research Society, 33, 63-71.
Kwahk, K. Y., & Kim, Y. G. (1999). Supporting business process redesign using causal
maps. Decision Support Systems, 25, 155-178.
Langfield-Smith, K. (1992). Exploring the need for a shared causal map. The Journal of
Management Studies, 29, 349-368.
Langfield-Smith, K., & Wirth, A. (1992). Measuring differences between causal maps.
The Journal of the Operational Research Society, 43, 1135-1150.
Lant, T. K., & Shapira, Z. (2001). Organizational cognition: Computation and interpretation.
Mahwah, NJ: Lawrence Erlbaum.
© 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Laukkanen, M. (1994). Comparative cause mapping of organizational cognitions. Organization
Science, 5, 322-343.
Lee, S., Courtney, J. F., & O’Keefe, R. M. (1992). A system for organizational learning
using causal maps. Omega, 20, 23-36.
Lee, S. Y. (1995). An integrated model of land use/transportation system performance:
System dynamics modeling approach. Ph.D. Dissertation, University of Maryland,
College Park.
Levi, A., & Telock, P. E. (1980). A cognitive analysis of Japan’s 1941 decision for war.
Journal of Conflict Resolution, 24, 195-211.
Linstone, H. A. (1984). Multiple perspectives for decision making: Bridging the gap
between analysis and action. New York: North-Holland, Elsevier Science.
Markoczy, L., & Goldberg, J. (1995). A method for eliciting and comparing causal maps.
Journal of Management, 21, 305-333.
Massey, A. P., & O’Keefe, R. (1993). Insights from attempts to validate a multi-attribute
model of problem definition quality. Decision Sciences, 24, 106-125.
Massey, A. P., & Wallace, W. A. (1996). Understanding and facilitating group problem
structuring and formulation: Mental representations, interaction, and representation
aids. Decision Support Systems ,17, 253-274.
Mitroff, I. I., & Linstone, H. A. (1993). The unbounded mind: Breaking the chains of
traditional business thinking. New York: Oxford University Press.
Montazemi, A. R., & Conrath, D. W. (1986). The use of causal mapping for information
requirements analysis. MIS Quarterly, 10, 44-55.
Nicolini, D. (1999). Comparing methods for mapping organizational cognition. Organization
Studies, 20, 833-861.
O’Keefe, J., & Nadel, L. (1978). The hippocampus as a causal map. Oxford, UK:
Clarendon.
Richardson, G. P. (1991). Feedback thought in social science and systems theory.
Philadelphia: University of Pennsylvania Press.
Sakman, M. (1985). An empirical study of three methods of problem definition in illstructured
situations. Ph.D. dissertation, University of Wisconsin, Madison.
Schneider, S., & Angelmar, R. (1993). Cognition in organizational analysis: Who’s
minding the store? Organization Studies 14, 352-374.
Senge, P. M. (1990). The fifth discipline: The art and practice of the learning organization.
New York: Doubleday/Currency.
Sheetz, S. D., Tegarden, D. P., Kozar, K. A., & Zigurs, I. (1994). A group support systems
approach to causal mapping. Journal of Management Information Systems, 11, 31-
57.
Steinbruner, J. D. (1974). The cybernetic theory of decision: New dimensions of political
analysis. Princeton, NJ: Princeton University Press.
Tegarden, D. P., & Sheetz, S. D. (2003). Group causal mapping: A methodology and
system for capturing and evaluating managerial and organizational cognition.
Omega, 31, 113-125.
Vennix, J. A. M. (1996). Group model building: Facilitating team learning using system
dynamic. Chichester, UK: Wiley.
Weick, K. E., & Bougon, M. G. (1986). Organizations as causal maps: Charting ways to
success and failure. In H. Sims & D. Gioia (Eds.), The Thinking Organization (pp.
102-135). San Francisco: Jossey-Bass.
Zhang, W. R., Wang, W., &King, R. S. (1994). A-pool: An agent-oriented open system
shell for distributed decision process modelling. Journal of Organizational
Computing, 4, 127-154.
Endnotes
1 In the analysis, we use their distance ratio (DR) formula to measure the distance
between collective maps and individual maps.
2 This framework and a prototype of sustainable decision support systems were
developed to improve policy planning and decision making regarding urban
infrastructure investments such as investments in roads and bridges, fresh water
supply systems, waste water treatment, drainage and so forth.
3 These interviews were conducted by the participants and other researchers with
people who are involved with the city’s infrastructure management.
4 It was our intention to limit the number of nodes to 15 (maximum).
5 The power-related implications of a map can be surfaced if it can be used to show
how one group of stakeholders is advantaged in the current situation and how this
contributes to the problems or how one group can manipulate the organization to
serve their interests.
6 The map, which had more links, thus, higher degree of map density ratio, received
a higher rating.