Comparing three simultaneous measurements of a sociocognitive network

Three data collection strategies are compared in their ability to measure a globally shared cognitive network representation, i.e. a sociocognitive network. Participant performances in data production using each of the three strategies are also assessed. Data were collected from 22 students in a graduate program. Each participant reported k out of 22 "ordinary" ties (Task 1, 0 less than or equal to k less than or equal to 22), in out of k "close" ties (Task 2, 0 less than or equal to m less than or equal to k), and "top three" ties (Task 3) for every actor in the network. Results revealed that each of the three tasks were quite similar to each other in measuring the sociocognitive network. The estimated average shared information among the actors about the network's structure, as revealed by each task, was highest for Task 2 and lowest for Task 1. Individual differences in performance indicate that the amount of shared knowledge as revealed by each task is not consistent across tasks. Overall, three-way data structures are shown to be sufficiently rich in information that any of the tasks are sufficient to capture the main features of a sociocognitive network. (C) 2002 Elsevier Science B.V. All rights reserved.