Cognitive Mapping

Conceptual Overview

Cognitive mapping techniques feature in studies spanning virtually every branch of management and organization studies, from marketing and strategic management, to technological innovation and human resource management. Two of the most popular techniques currently in use are repertory grid and causal mapping. The former exemplifies a range of approaches suitable for revealing dimensional and hierarchical relationships among conceptual stimuli, the latter a class of techniques for investigating actors' conceptions of influence, causality, and system dynamics.

Repertory grid procedures originated in the field of clinical psychology, based on the personal construct theory of G.A. Kelly. This theory asserts that individuals behave in a manner akin to natural scientists as they go about their everyday business, formulating hypotheses about their worlds, which they then seek to “validate.” Basic judgments are informed by cognitive maps in which the stimuli encountered (elements) are organized along a series of bipolar dimensions (personal constructs). In the classic, minimum context form of the repertory grid technique, participants' personal constructs are elicited using a triadic elicitation procedure. The elements to be mapped are presented at random on a series of cards. Participants are required to explain how a given pair of elements are alike but different from a third. This task is repeated until an exhaustive list of constructs has been revealed. For example, in a job analysis exercise the elements might comprise the main day-to-day tasks (e.g., chairing meetings, conducting interviews, driving). Applying the above procedure would yield a range of constructs (e.g., “safe vs. risky task,” “little contact with people vs. high contact with people”). Once a full set of constructs has been elicited, these can be converted into a series of bipolar rating scales, and the participant is required to evaluate all the elements in terms of all of the constructs. The resulting quantitative data can be submitted to multivariate analysis techniques (e.g., principal components analysis, multidimensional scaling, and cluster analysis) to drive visual representations depicting the interrelationships among the elements and constructs.

An emphasis on action, focusing on the perceived interrelationships between a given situation and its antecedents and likely consequences, renders causal mapping approaches particularly attractive. In its earliest form, causal mapping involved the coding of documentary sources and interview transcripts. In recent years, however, causal mapping techniques have been adapted in order to study actors' knowledge structures in a relatively direct fashion, through face-to-face interviews and structured questionnaires. Regardless of how the data are gathered, the end result typically takes the form of an influence diagram, in which causal assertions between a set of variables are depicted as a network of associations interconnected by arrow-headed pathways, terminating in each case on the dependent variables, the arrowheads depicting the directions of perceived causality. Relationships can be positive (an increase in one variable is believed to cause a corresponding increase in the dependent variable) or negative (as one variable increases, the other decreases). [Page 190]More sophisticated variants of the technique enable pathways to be differentially weighted in accordance with varying belief strengths and permit the representation of reciprocal relationships (see Figure 1). As with repertory grids, larger, more sophisticated maps are best submitted to formal, quantitative analyses, using algebraic techniques.

...