Search

We want to do three things in this paper: (1) to propose a more general probability model than the one considered by Toby or Davies for dealing with the number of buses a student would see, on the average, going in the wrong direction; (2) to show that, with this general model, one would not see more wrong-direction buses, or the same number of wrong-direction buses, as suggested by Toby and Davies, but fewer wrong-direction buses; (8) to question the advisability of using a student's experiences with wrong-direction buses as the way to shake his faith in the validity of personal experience. As Davies has said, clarification is especially necessary now that Toby's paper has appeared in books of readings designed for the course in introductory sociology. [ABSTRACT FROM AUTHOR]

The article describes a process for determining migration probability. A stochastic process may be defined as a process which develops over time and where events at time 2 are not independent of events at time 1. This lack of independence may be viewed in probabilistic terms. Since the application of stochastic probability processes to problems of predicting individual behavior requires data obtained from case studies where each individual can be classified with respect to a given set of categories at successive points in time, the projected analysis required a set of individual longitudinal migration histories. Consequences of the basic mechanism of the model utilized in the paper are also detailed.

Science seems to be, among other things, part of man's quest for legitimate extrapolations beyond immediate sense impressions. Though some sociologists seem to think that "going beyond one's data" constitutes a scholarly sort of sin, the whole purpose of a sampling apparatus is precisely to enable a person to make claims regarding parameters from statistical knowledge. Further, the development of scientific theory appears to be of the same order. Science is not a mere cataloging of either true or verifiable statements. Scientific statements cohere in an organized body of interconnected propositions. The basic scientific act is to make a judgment regarding the acceptability of these statements (Kaufman, 1944; esp 39ff). Since a variety of different kinds of statements can be identified, the grounds for acceptance vary according to their nature (Francis, 1969). Further, criteria such as "elegance" and "pragmatic workability," though creating a certain degree of tension at times, rest on different grounds than, say, establishing a probability level for Type I statistical error. In any event, the strain is toward consistency such that neither the statement itself nor the grounds for accepting it are contradicted elsewhere. [ABSTRACT FROM AUTHOR]

How can you use a product-moment correlation coefficient when you do not have interval scale data? This is particularly addressed to uses of such statistics with personality, value, or other 'soft' variables, and to variables that are sometimes considered to be composed of discrete categories. The implicit contrast usually is to a nonparametric or sidtribution free statistic like Maurice G. Kendall's Tau as more appropriate with such data. Since part of the confusion that surrounds the question of the appropriateness of use of various statistics occurs about the type of measurement involved, we may focus on the presentation of Herbert M. Blalock on this topic.