Identifying Objects
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
Figure 5.Construct level causal map
Behavior
Structure
Object Oriented
Development
Systems
Object Oriented
Modeling
The individual maps are then aggregated (Axelrod, 1976; Bougon et al., 1977). The
aggregation is performed at both the concept level and the construct level. The
aggregation process consists of combining the causal maps of each participant into a
single aggregate map.1
Validation of Maps
Once the maps have been created, they should be validated. The validation method is
determined by the data collection method (IECM or TBCM). As a source of validation
for an IECM a member check may then be performed using the aggregated maps to ensure
accurate and comprehensive representation (Lincoln & Guba, 1985). Each participant (or
as many as you have access to) is shown the aggregated maps and asked if the maps
accurately reflect the concepts, linkages and constructs. It should not be unexpected that
as the participant walks through the map he or she will be surprised. The map reflects
multiple causal relationships and most individuals do not consciously perceive the
causality of concepts in terms of a network. The key is to engage the participant so he
or she can reflect on the map you constructed based on the interviews. After a thorough
discussion of the map with the participant, any discrepancies should be reported. For
example, Nelson et al. (2000) fed the maps back to the organization to get feedback on
maps.
When using TBCMs (e.g., archival data) validation becomes a more complicated process
because there is no one to confirm your results. With TBCMs one commonly used method
of validation is triangulation with other sources. For example, if the researcher is using
change request data to track software development productivity, additional data may be
gathered from departmental annual reports or individual annual reviews. In another
example, Nadkarni and Narayanan (in press) validated the causal maps they constructed
from annual reports of firms with the firms K-10 statements. Both internal and external
sources can be used if available.