Shows how neural networks can bring together psychometric and econometric approaches to the measurement of attitudes and perceptions. Uses a neural network to analyse data collected from a sample of ATM users on their perceptions of ATM service. Uses the weights of connections from input nodes to hidden nodes to label the hidden nodes to represent particular respondent attitudes. Uses the network to analyse the impact of explanatory (input layer) variables on the hidden layer attributes, and through these on the endogenous (output layer) variables ‐ satisfaction with ATMs, likelihood of recommendation to others, extent and frequency of use. Defines four user types, characterized as “disaffected youth”, “technophobes”, the “pro‐technology” segment, and the “cost conscious” segment. Gives some ideas on how banks could address the needs of each segment.
CitationDownload as .RIS
MCB UP Ltd
Copyright © 1996, MCB UP Limited