On testing the equality of proportions in the paired case with incomplete data

The problem of testing two correlated proportions with incomplete data is considered by means of Monte Carlo simulations studies. A test proposed in this paper, which can be regarded as a generalization of McNemar's test, is recommended in all cases with incomplete data and not too small samples.     

A multidimensional measure of neighboring

The study reports the development of a short and easily administered questionnaire aiming at measuring dimensions of social life within neighborhoods. Principal-components analysis consistently extracted four factors replicated in three independent samples (N=96 to 1,060). The factors emerged as theoretically meaningful dimensions tapping the concepts of supportive acts of neighboring, neighbor annoyance, neighborhood attachment, and weak social ties. Factor invariance and factor replicability were high. Internal consistency, test-retest reliability, and construct validity yielded acceptable results. The principal virtue of the measure is that it may be applied to discern qualitative differences between neighborhoods by simultaneous assessments of several dimensions of neighboring. This research was financially supported by The Norwegian National Research Council. We thank Jostein Rise for valuable comments, and David R. Jacobs, Jr., and Melanie Young for advice in translating the questionnaire.     

The tower experiment and the Copernican revolution

During the Copernican revolution the supporters of the Ptolemaic theory argued that the tower experiment refuted the Copernican hypothesis of the (diurnal) motion of the earth, but was in agreement with the Ptolemaic theory. In his defence of the Copernican theory Galileo argued that the experiment was in agreement both with Copernican and Ptolemaic theory. The reason for these different views of the same experiment was not that the two theories were incommensurable, as Paul Feyerabend argues, but that Galileo introduced a new theory of motion which he used as an auxiliary hypothesis in his discussion of the tower experiment, while those defending the Ptolemaic theory used the old Aristotelian theory of motion. Already before the Copernican revolution the Aristotelian theory of motion was criticized by philosophers in Paris, who suggested the impetus theory of motion. The later versions of this theory had the consequence that the tower experiment no longer refuted the hypothesis of the (diurnal) motion of the earth. Thus the impetus theory removed an old and important objection to the heliocentric theory. Galileo's inertial dynamics had the same function in the discussion of the tower experiment.     

PERCEPTION OF MOTION AND CHANGING FORM: A study of visual perception from continuous transformations of a solid angle of light at the eye

It is shown how geometrically changing projections of objects which move and/or change their shape carry no specific information about form and three-dimensional motion. How, then, does the visual apparatus produce specific percepts from such non-specific changing stimuli? By applying an analogue computer technique, changing projections of artificial objects are generated on a CRT screen. These projections are fed into the eye by means of an optical device where they form a continuously changing solid angle of homogeneous light. The main conclusion is that it is a principle of perceptual three-dimensionality which gives specificity to the percepts. Preliminary statements of principles for prediction of perceived motion in depth from a given change in proximal stimulus are presented.     

DIFFERENTIATION OF CHARACTER NEUROSIS AND SYMPTOM NEUROSIS ON THE BASIS OF DIFFERENCES BETWEEN TWO SERIAL EXPERIMENTS

The difference between two serial tests as regards signs of hysteria and obsession-compulsion is compared with differences in number of symptoms reported by the patient. Signs in the more formal colour-word test, measuring habitual behaviour patterns, correspond to fewer symptoms than signs in a tachistoscopic test revealing more of the perceptual representation (content) of behaviour disorders. This difference is supposed to reflect important aspects of the difference between character neurosis and symptom neurosis.     

THE STABILITY OF PATHOLOGIC SIGNS IN THE PERCEPTUAL PROCESS AS REVEALED BY A SERIAL TACHISTOSCOPIC EXPERIMENT

S mith , G. J. W., and J ohnson , G. The stability of pathologic signs in the perceptual process as revealed by a serial tachistoscopic experiment. Scand. J. Psychol ., 1961, 2, 142–148.—The study demonstrates high inter-rater and inter-set consistency in a serial tachistoscopic experiment utilizing pairwise presentations of stimuli A and B, B being presented at constant time levels, A at gradually increasing ones. Two parallel sets contain two series each. In one series A is incongruent with B, in another a threat directed against B. Fifty psychiatric patients are given the two sets with a one-day interval. Reports of change in A and B during the micro-genesis of A are scored in the dimensions of repression, isolation, projection, and depression.     

BINOCULAR INTERACTION IN MOTION PERCEPTION

An earlier experiment has shown a high degree of overestimation of small motion tracks as compared with static space perception. In the present paper this is hypothesized as the visual outcome of a central nerve process specific for motion stimulation. The hypothesis is submitted to a test in an attempt to show a possible binocular intereaction of the effect. The experiment demonstrated a significant binocular summation and thus supported the hypothesis.