Iterative least squares estimates of communality: Initial estimate need not affect stabilized value

A common criticism of iterative least squares estimates of communality is that method of initial estimation may influence stabilized values. As little systematic research on this topic has been performed, the criticism appears to be based on cumulated experience with empirical data sets. In the present paper, two studies are reported in which four types of initial estimate (unities, squared multiple correlations, highestr, and zeroes) and four levels of convergence criterion were employed using four widely available computer packages (BMDP, SAS, SPSS, and SOUPAC). The results suggest that initial estimates have no effect on stabilized communality estimates when a stringent criterion for convergence is used, whereas initial estimates appear to affect stabilized values employing rather gross convergence criteria. There were no differences among the four computer packages for matrices without Heywood cases.     

Geophysical variables and behavior: XXIX. Impact of atmospheric conditions on occurrences of individual violence among Canadian penitentiary populations

This project studied the effects of weather variables and size of the population on minor and major violence rates within six prisons in the Kingston, Ontario, Canada area from January, 1980 through December, 1983. While weather variables have been studied in relation to mood changes within the general population, this study was the first to analyze the impact of weather variables on rates of violence within the prison setting using forward inclusion multiple regression. Analysis showed that population size consistently appeared positively correlated with assaults on other inmates among the male medium-security prisoners during the summer months. In the summer months, temperature tended to be positively related to some minor kinds of offenses, but in the winter months the relationship was negative. Humidity, rain, sunshine hours, snow, and snow on the ground were not consistently related to incidences of minor and major violence. Indices of geomagnetic disturbances had statistically significant inverse relationships with attempted suicide/self-inflicted injury rates among the male prisoners during the summer months. Over 12 months, wind was generally negatively correlated with incidences of major and minor violence among the male inmates. For 6-mo. periods, wind was generally positively correlated with violence rates during the winter months and negatively correlated during the summer months among the male inmates. The findings relating weather variables to violence rates in the women's prison were consistently in opposition to those found for male prisoners during the 4-yr. period.     

A reflection: Howard Scott Liddell, 1895-1962

Howard S. Liddell, Professor of Psychobiology in the Department of Psychology at Cornell University, was one of the earliest American students and developers of I.P. Pavlov's theory of the conditioned reflex and an early researcher on stress. The significance of Liddell's work on stress was fully recognized when he was sent, as a member of a commission, to the battlefront at the height of the Korean War to study stress and self-control in the soldiers. This profound experience convinced Liddell that behavioral research must ultimately be made relevant to the understanding and elevation of human behavior. He spent the remaining ten years of his life in efforts to implement this conviction through research, writing, and lectures before scientific and lay groups.     

A logic of message and reply

In this paper we sketch a logic of message and reply. The logic is intended for application in a wide variety of situations, not restricted to the two-person, turn-taking situation. Each message has a body and a vector; the vector specifies the from, to, and the like. To reply to a message, it suffices to give either (1) a complete reply to the body or (2) a corrective reply to at least one presumption derivable from the vector. We discuss the problems of achieving effectiveness and completeness with respect to certain aspects of communication. The results are mixed. In section 9 we argue semi-formally that, in a certain sense, dialogue is necessary. Finally we note that this logic is not a rival of other approaches but may be combinable with them.