Variance Index Research Articles

A computer program for the generation of Hudson’s variance index for categorical clustering performance

A computer program for the generation of Hudson’s variance index for categorical clustering performance

Between groups and total income inequality indices; U.S. 1960∗

The study contains a theoretical analysis on the behavior of a between groups Lorenz Index. The variables affecting this index are utilized to construct between groups indices for white-non-white; rural-urban; above 65 yr old-younger; population groups by state for 1960. The constructed indices were then used as explanatory variables for the state total Lorenz Index. It was found that they explained over 80 per cent of the variance. Adding other socio-economic variables enabled us to explain over 90 per cent of the variance of state Lorenz indices.

A comparative survey of the results of analyses of blood serum in clinical chemistry laboratories in the United Kingdom.

Since July 1969, portions of the same blood serum have been dispatched to clinical chemistry laboratories in the United Kingdom at 14-day intervals. The results of each serum survey were reported to each of the 390 participants within 11 days of their originally receiving the specimen. During the first 18 months of the survey no overall improvement in the results was seen. Therefore a summary of each laboratory's ability consistently to produce results close to the mean of the method used was calculated and reported as a single figure, the variance index, and sent to all participants at regular intervals together with a histogram distribution of the variance indices of other participants. The subsequent improvement in the overall results is described.

EEG variation in healthy subjects

1. 1. An analysis of variance of certain EEG indices was performed on 27 healthy subjects. 2. 2. Simple visual procedures were performed by each subject in order to maintain a constant level of attention. It was found that no particular procedure or alteration of the order of presentation had a greater stabilizing effect. 3. 3. Differences of age and sex had no effect on the variances of the EEG indices studied, nor did it seem necessary to allow for the effects of menstruation on the EEG. 4. 4. Marked fluctuation during the course of a single EEG recording usually foretold marked fluctuation from one day to the next. 5. 5. As the data did not conform to a normal distribution it was difficult to express EEG index variability in an orthodox manner. An ad hoc technique is described for this problem.

断続音と白色雑音の, 精神作業及び感情への影響

This investigation was carried out to know whether the same kind of results as previously reported (Sato, 1972) could be obtained or not, even when the nature of the stimuli changes from flickering light to intermittent sound and white noise. Results: 1) The curve, in which variance indices of the correct responses of the task given Ss were plotted against the frequency of flutter, showed two peaks for all Ss. The one peak was at 2Hz, and the other at 12Hz. In the flickering light exp. previously reported, the Ss were divided into two groups, each group having only one peak on this curve. 2) All of the flutter and noise stimuli were estimated to be unpleasant on pleasant-unpleasant rating scales by the majority of Ss. But small number of Ss estimated them to be pleasant in every frequency. 3) The relationship between response variability of the task and the feeling caused by the flutter and the noise was only slight.

FURTHER RESULTS ON THE STANDARD ERRORS OF ESTIMATE ASSOCIATED WITH ITEM‐EXAMINEE SAMPLING PROCEDURES

Defining one observation as the score received by one examinee on one item, the results of this investigation suggest that, for a given test length, item‐examinee sampling procedures having the same number of observation have, for all practical purposes, the same standard error in estimating μ but different standard errors in estimating σ. Additionally, the variance of the item difficulty indices (proportion answering the item correctly) was found to be a significant factor in accounting for differences in standard errors of estimating μ between normative distributions differing primarily in degree of skewness.

Effects of Task Structure on the Process of Group Organization in Terms of the Difficulty of the Task: I

The task was a kind of color-naming task which was solved by making seven pairs of seven given colors and ten given adjectives. The criterion for task completion was simply a matter of group consensus. The level of difficulty was defined as the degree of attainability of consensus in naming. The difficult task and the easy task were determined from pilot experimentation. Six groups, each composed of three 10- to 11-yr.-old boys, were Ss. Each group performed cooperatively the difficult and the easy tasks. Discussion in each group during the process of reaching a decision was syste~natically observed and analyzed according to the categories which have been described in derail elsewhere (Nagata, 1965). The main results may be summarized. The easy task induced more groupsteering (leadership ) role and task-role differentiation than did the dif ficult task. The result of analysis of variance of mean indices (Dickens, 1955) for the role differentiation based on arc sine transformed scores, showed that the difficulty of the task was significant (F = 9.288, df = 1/5, p < .05). Also, the difficult task induced a more passive and dependent style of steering activity (leadership style) on the part of leaders than did the easy task. In this case, the leader was defined as the one who showed most frequently those activities classified as the group-steering activities.

Signal variance and its application to continuous measurements of e.e.g. activity

This analytical technique provides numerical values which may be used as objective measures of e.e.g. activities. The time scale of the analysis may be so chosen that an e.e.g. record of any duration can be represented by a single graph, on which to the same scale, other associated physical or physiological variables may be recorded. A means is provided for relating time points on the original record with the same time points in the analysis. The method is illustrated by analyses of e.e.g. changes ( a ) brought about by anoxia following rapid decompression, ( b ) during sleep, and ( c ) following the injection of pentothal and the insertion of sphenoidal electrodes. It has also been used in studies of hyperventilation, changes resulting from increased gravitational stress on a human centrifuge, and of the effect of anaesthetics on arousal patterns in animals. The e.e.g. signals are divided by filters into five frequency bands approximating the accepted physiological definitions. For each band the signal variance is calculated and the growth of all five variance integrals displayed simultaneously as a set of continuous curves, the slopes of which are measures of activity in the bands. The ratio of any two slopes—the ' variance index’ — may be used as a measure of change between activities in one band at different times, between different bands of the same recording, or between different recordings. The same curves may be used to produce frequency spectra in cases where this approach would be justified: changes in the slope of the variance integral curves will determine the validity of the spectral analysis approach. All the operations are carried out on a simple general purpose analogue computer and the results recorded on an X Y plotting table or multichannel mirror galvanometer recorder. E.e.g. signals may be analyzed ‘on line’, i.e. as they are recorded, or, by means of a magnetic tape recorder, specific portions of the record can be examined later in greater time or amplitude detail.

Coeficiente de variação e índice de variança

The object of the present paper is a study of the usefulness of two relative measures of variation, the well known coefficient oj variation and a new term proposed in this paper and called the index of variance. These terms are defined by the equations : coefficient of variation : σ% = s /v/v. 100 index of variance :σ/ √. 1. It is shown that, for theoretical reasons, only the index of variance may be expected to be constant. Six different experimental series actually proved this constancy, showing at the same time the variability of the coefficient of variation which proved to be dependent upon the respective mean. The coefficient of variation becomes approximately constant when the respective means are sufficiently distant from the absolute limit zero or other biological limits. 2. Thus the index of variance may be used to prove the homogeneity of variation in samples with means of different dimensions. 3. Through this it is shown that the index of variance should be constant, it is explained that for biological reasons we may not always find a good fit between the observed and the expected data, calculated by means of the equation = kv where k represents a biological constant. In two cases a better fit was obtained using more complicated formulae. 4. While it seems justified in agricultural experimentation to accept proportionality between mean yield and area, no such relation exists for the standard error. We may only say that generally the index of variance for large areas is not equal, but bigger than that for smaller ones. 5. The coefficient of variation cannot be used as a general term for comparing the variation in series of different dimensions, where we must apply the index of variance. But it still retains its value as a measure of the efficiency of experiments.