Cognition

К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
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 

When maps have been elicited using highly structured rating scales common to each

participant, it is possible to construct one or more aggregate maps by combining

individual adjacency matrices. At the most basic level such aggregation can be accomplished

by simply adding or averaging participants’ judgments of common causal

relations (e.g., Bougon, Weick & Binkhorst, 1977; Ford & Hegarty, 1984; Voyer &

Faulkner, 1989).

Aggregated maps do not necessarily reflect the views of any one individual. However,

they are potentially insightful insofar as they enable the detection of overall group

tendencies, the possibility of widespread within-group variance notwithstanding (cf.,

Walsh, 1995). The latter, of course, is detectable by computing basic measures of spread,

such as the semi-inter-quartile ranges or standard deviations associated with particular

mean responses. Such aggregate analyses can be highly insightful, as, for example, in

the identification of mean sub-group differences in the perception of how much a given

construct influences, or is influenced by, other constructs, and its overall detrimental or

beneficial effect. Aggregate causal mapping methods also permit the study of the overall

structure of group-level mean perceptions of a given set of constructs, thus extending

the analysis beyond such basic bivariate relationships.

Table 2. Four indicators reflecting potentially significant differences between cause

maps (after Langfield-Smith & Wirth, 1992)

Difference

Meaning

1 The existence vs. non-existence of

particular variables

One individual or group believes that a particular

variable is important, whereas a second individual

or group does not

2 The existence vs. non-existence of

relationships between particular variables

One individual or group believes a given variable

has an influence upon or is influenced by another

variable, whereas a second individual or group

does not

3 The polarity of relationships represented

within the maps

One individual or group believes that the

relationship between two given constructs is

negative, whereas a second individual or group

believes the relationship is positive

4 The polarity strength

Two individuals or groups hold the same belief

about the polarity of a given relationship but one of

them believes the relationship is stronger than the

other