The Use of Regression on Dummy Variables in Management Research

Studies of organizational behavior and other managerial phenomena frequently include variables which are not readily measurable by cardinal or ratio scales. Variables of this nature occur, for example, when investigating the relationship between a supervisor's span of control and (i) dissimilarity of subordinate's jobs, (ii) supervision received from other executives, (iii) geographic concentration of subordinates, and (iv) personal assistants. The (0, 1) variable technique, explained and illustrated here in terms of this example, offers the researcher a superior analysis for those sociological, psychological, and managerial phenomena for which no metric has yet been developed. The authors believe that this is one of the most useful statistical techniques available to the management scientist.

• 저널Management Science · 16(6) · B–397
• 토픽Advanced Statistical Methods and Models · Statistics and Probability
• DOI10.1287/mnsc.16.6.b397
• 저자S. R. Searle, Jon G. Udell