Bayesian statistics and biased procedures

A comparison of Neyman's theory of interval estimation with the corresponding subjective Bayesian theory of credible intervals shows that the Bayesian approach to the estimation of statistical parameters allows experimental procedures which, from the orthodox objective viewpoint, are clearly biased and clearly inadmissible. This demonstrated methodological difference focuses attention on the key difference in the two general theories, namely, that the orthodox theory is supposed to provide a known average frequency of successful estimates, whereas the Bayesian account provides only a coherent ordering of degrees of belief and a subsequent maximization of subjective expected utilities. To rebut the charge of allowing biased procedures, the Bayesian must attack the foundations of orthodox, objectivist methods. Two apparently popular avenues of attack are briefly considered and found wanting. The first is that orthodox methods fail to apply to the single case. The second is that orthodox methods are subject to a typical Humean regress. The conclusion is that orthodox objectivist methods remain viable in the face of the subjective Bayesian alternative — at least with respect to the problem of statistical estimation.     

Testing significance of differences between two chi-squares

This paper presents a way to test the difference between twoX ²'s. The test requires evaluating the difference with respect to theT _m (x) Bessel function. Included is a table of the 5 percent and 1 percent points for the Bessel function with degrees of freedom up to 100.This investigation was supported by National Science Foundation Grant Number GE 9846.     

The Vocational Rehabilitation Counselor on Mental Patient Treatment Teams

This study asked the question: What are the role expectancies of treatment personnel in the treatment community and are they consistent across patients' emotional condition? An 83-item inventory was developed and administered under three different sets of instructions, each characterizing a different patient condition. Treatment staff and administrators were asked to rate items on the basis of the level of expectancy they held for their team vocational rehabilitation counselor. The analyses suggest that treatment staff and counselors do not agree on role expectations and that patient condition is a significant mediating variable.     

Reading and typing

As an extension to an earlier study on typing, an experiment was carried out on skilled touch typists in which they were asked to type texts that differed in word structure and word length. It was predicted from certain assumptions about the reading process that there should be significant effects of structure, length and their interaction. These predictions were confirmed.     

Are thresholds reduced by illusions? An attempt at replication

Three experiments are reported, which are attempts to replicate the finding of Ross and Gregory (1964) that difference thresholds for weights can be lowered by means of the size-weight illusion. Three different procedures were used, the first one (experiment I) being designed to show whether or not changes in a subject's judgement criterion could account for apparent changes in sensitivity. The second method (Experiment II) was a replication of Ross and Gregory's first procedure, in which the standard weight was judged before the comparison. In Experiments I and II a larger illusion was induced than in the original studies, but in Experiment III both the weights and container sizes were practically identical to those used by Ross and Gregory. The procedure was also the same as their most successful procedure (number 3) in which standard and comparison were judged simultaneously. The findings were uniformly negative: there was no evidence of criterion shift when the size-weight illusion was induced nor did we find the lowering of threshold previously reported.     

The bioelectrical activity of cerebellar structures in the freely-moving rat

Electrodes were implanted in the brains of 27 freelymoving rats and the bioelectrical activity of cerebellar cortical structures (lobus simplex, tuber verrais, lobus ansiformis, crus II) and the dentate nucleus was measured simultaneously with the activity of the cerebral cortex and dorsal hippocampus and respiratory rate and motor activity. Different behavioral states were produced by habituation procedures and by elaborating conditional avoidance reflexes to light-flash or click series. In addition, startle reflexes to acoustic stimuli were used to evaluate behavioral state. The following conclusions could be drawn: 1) in the awake rat the various cerebellar structures have clearly distinguishable bioelectrical activity patterns; 2) changes in these patterns depend on the actual behavioral state of the animal; 3) this dependence upon behavior is especially clear in crus II, one of the projection zones of the tactile and proprioceptive afferent nerves in the cerebellar cortex. The changes in the electrocerebellogram of the unrestrained rat may be used as an indicator of the behavioral state of the animal.