A Markov chain model for Asch‐type experiments†

The data used in this study were collected by Bernard P. Cohen, Department of Sociology, Stanford University. Financial support for the research being reported was provided by the following: three National Science Foundation grants ('Probability Models for Conformity,’ Bernard P. Cohen, Principal Investigator; ‘Authority Structures and Evaluation Processes,’ Stanford M. Dornbusch, Principal Investigator; ‘A Comparative Study of Models for Binary Data,’ Hans E. Lee, Principal Investigator); grants of computer time from the Computer Center, Stanford University; grants and computer time from Michigan State University. The author was supported by the Alice Rosenberg Research Fellowship, by a Stanford Graduate School Fellowship, and by a Traineeship awarded to the Department of Sociology, Stanford University, by the National Institutes of Health. The author acknowledges his indebtedness to these sources of support and is happy to have this opportunity to thank each of these individuals and agencies.

A 4‐state absorbing Markov chain is advanced as a model to represent the response process of subjects in Asch‐type experiments. Both the development of the model and the estimation of its parameters are described. The model is applied to the responses obtained from subjects in an Asch‐type experiment designed to analyze the effects of prior status expectations on social influence.

Notes

The data used in this study were collected by Bernard P. Cohen, Department of Sociology, Stanford University. Financial support for the research being reported was provided by the following: three National Science Foundation grants ('Probability Models for Conformity,’ Bernard P. Cohen, Principal Investigator; ‘Authority Structures and Evaluation Processes,’ Stanford M. Dornbusch, Principal Investigator; ‘A Comparative Study of Models for Binary Data,’ Hans E. Lee, Principal Investigator); grants of computer time from the Computer Center, Stanford University; grants and computer time from Michigan State University. The author was supported by the Alice Rosenberg Research Fellowship, by a Stanford Graduate School Fellowship, and by a Traineeship awarded to the Department of Sociology, Stanford University, by the National Institutes of Health. The author acknowledges his indebtedness to these sources of support and is happy to have this opportunity to thank each of these individuals and agencies.