Causal Mapping Background
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
Causal maps have been used to represent managerial cognition at both the individual and
group levels (Axelrod, 1976; Eden & Ackermann, 1998a; Huff, 1990; Meindl, Stubbart, &
Porac, 1996). From a managerial and organizational cognition perspective, five causal
mapping approaches have been used to produce collective causal maps (see Table 1).
Most collective causal map approaches capture the data for the collective maps using
individual maps. The individual maps tend to be either created using a participant-driven
interview, such as the Self Q interview (Bougon, 1983), or a negotiated researcher and
participant interview (Eden & Ackermann, 1998a). The advantage of a participant-driven
approach is the minimization of the possibility of researcher bias impacting the creation
of the individual maps (Nicolini, 1999).
All of the approaches for creating collective maps from individual maps require that the
concepts used in individual maps be standardized in order to create collective maps. The
use of congregate labels created by the researcher to group similar concepts used across
individuals is common to all approaches that merge individual maps into collective maps
(Bougon, 1992; Eden & Ackermann, 1998a, 1998b). In the merging processes associated
with the first four approaches in Table 1 (congregate, shared, group, and oval maps), this
standardization process occurs after the individual maps are created. Congregate labels
are based on researcher’s and possibly participant’s identification of similarities of
beliefs contained in the individual maps. The congregate labels are then substituted in
the individual maps. Once the congregate labels have been placed into the individual
maps, the individual maps can then be merged based on the common nodes (congregate
labels) contained in the individual maps. As a result, the process of merging individual
maps into a collective map is both time consuming and results in a loss of information
regarding idiosyncratic differences among individual belief structures. In addition,
researcher bias may be present as the research/facilitator usually determines the
congregate labels across individual maps (Nicolini, 1999). In contrast, our process
(Tegarden & Sheetz, 2003) enables the individuals in a decision-making team to agree
upon the congregate labels so that researcher intervention and bias is minimized.
Type Data Capture Approach Data Merging Approach
Congregate Map
(Bougon, 1992)
Participant-Driven Researcher-Driven
Shared Map
(Langfield-Smith, 1992)
Researcher- and Participant-Driven Researcher- and Participant-
Driven
Group Map
(Eden & Ackermann, 1998a)
Researcher- and Participant-Driven Researcher-Driven
Oval Map
(Eden & Ackermann, 1998a)
Researcher- and Participant-Driven Researcher- and Participant-
Driven
Group Map
(Tegarden & Sheetz, 2003)
Participant-Driven Participant-Driven
Table 1. Types of collective cause maps
To identify cognitive diversity or what we call cognitive factions in a top management
team, we cluster the causal maps created by the members of the top management team
based on the similarity of the cause-effect linkages between the nodes (congregating
labels). A similar approach was employed by Reger and Huff (1993) to compare cognitive
similarities and differences of industry maps across top managers within an industry. In
the next section we describe the methodology used to identify the cognitive factions.