Search

The effects of the form of the function relating criterion values to cue values and of the validity of the cue upon the subjects' hypotheses in probabilistic inference tasks were studied in a factorial experiment with two levels of cue validity (rCE = A5, and rCE = .90), and four function forms (positive linear, negative linear, inversely U-shaped, and U-shaped). The results were consistent with a hypothesis sampling theory, and the relative frequencies of four basic hypotheses could be predicted from earlier results with respect to the relative strengths of those hypotheses. The results also showed that the correlation between the hypotheses stated by the subjects and the rules extracted from their judgments by means of polynomial regression was higher for linear hypotheses than for nonlinear hypotheses. The correlation was quite low, however, also for linear hypotheses. [ABSTRACT FROM AUTHOR]

Discusses Charles Peirce's systems of existential graphs known as systems of logic. Family of logical systems developed by Peirce; Reason why Peirce went through the effort of developing the graphs; Concept of mathematica reasoning.

The article focuses on the problem of prediction in sociological research. Perhaps the most inviting current methodological problems inherent in use of case materials in sociological research rest in attempts to extend the base for systematically derived generalizations from individual cases to classes of cases. In the idiographic frame of reference the ends of scientific generalization are understanding and powers of prediction and control pertinent to the behavior of a single case; in the nomothetic frame of reference the ends of scientific generalization are understanding and powers of prediction and control pertinent to cases grouped categorically. Some argue that actuarial predictions concerning given populations do not reveal which intra-class individual will be involved in predicted behavior. The chance of successful prediction relative to an individual's behavior, whether this is recognized or not, is dependent upon logic and manipulation identical with those relative to prediction by classes.

In some inference situations a report of an event must he distinguished from the event itself. Such a distinction is necessary whenever there is reason to question the reliability of the source which generates the report. A quantity called adjusted likelihood ratio (Λ) is defined which incorporates information about source reliability and information about the inferential impact (L) of the event being reported. A is shown to differ from L under several different conditions of source reliability. When source reliability is not contingent upon the truth of any hypothesis being considered, a small reduction in reliability drastically degrades very large values of L. When L is small to begin with the effects of reducing reliability are less drastic. When source reliability is contingent upon some hypothesis. then Λ can he greater than L for the event. [ABSTRACT FROM AUTHOR]