Influential observations in principal component analysis:a case study

A number of results have been derived recently concerning the influence of individual observations in a principal component analysis. Some of these results, particularly those based on the correlation matrix, are applied to data consisting of seven anatomical measurements on students. The data have a correlation structure which is fairly typical of many found in allometry. This case study shows that theoretical influence functions often provide good estimates of the actual changes observed when individual observations are deleted from a principal component analysis. Different observations may be influential for different aspects of the principal component analysis (coefficients, variances and scores of principal components); these differences, and the distinction between outlying and influential observations are discussed in the context of the case study. A number of other complications, such as switching and rotation of principal components when an observation is deleted, are also illustrated.     

Disaster and deritualization: A re-interpretation of findings from early disaster research

This study investigates how ritual practices are disrupted in disasters and the ways people deal with those situations. We employ structural ritualization theory to conduct this investigation focusing on deritualization, which refers to the breakdown or loss of ritualized activities that occur in daily life. The thesis of this paper is that ritual practices are disrupted in a disaster leading to the breakdown of action and meaning. Data were collected from 19 seminal sociological studies from the National Academy of Sciences published first under the name of the Committee on Disaster Studies (1951–1957) and then under the rubric of the Disaster Research Group (1957–1962). A content analysis of these studies revealed many instances of deritualization. Deritualization is manifested or most evident through its effects on action and meaning in rituals. Examples were categorized according to the loss of action, meaning, or a combination of both.     

Monte carlo evidence on the robustness of conditional moment tests in tobit and probit models

This paper numerically examines the size robustness of various conditional moment tests in misspecified tobit and probit models. The misspecifications considered include the incorrect exclusion of regressors, ignored heteroskedasticity and false distributional assumptions. An important feature of the experimental design is that it is based on an existing empirical study and is more realistic than many simulation studies. The tests are seen to have mixed performance depending on both the original null hypothesis being tested and type of misspecification encountered.     

Euler buckling in a wheelbarrow obstacle course: a catastrophe with complex lag

This article studies human physical work capacity under increasing load from a general systems theory perspective. There are several points of generalization between the structure of nonliving (building materials) and living systems (humans) with respect to stress, strain, and fatigue. A catastrophe model for Euler buckling was transposed and tested for human performance in a wheelbarrow obstacle course under varying loads. Subjects were 129 employees of a Midwest manufacturing plant. A cusp model was hypothesized and verified (R2 = .68, control R2 = .11) where vertical load was the asymmetry factor, and body balance, height, and sex-related differences all contributed to bifurcation. A catastrophe model in codimension 10 was also invoked to explain memory in the system. Principal control variables were exercise habits, weight, balance, and sex-related differences (R2 = .75). The core model of human load-to-failure was concluded to be similar to that for Euler buckling: additional complexities were discovered which were attributed in part to systemic memory. Discussion points included the use of large dimension catastrophe models for problems involving complex lag effects, and the transposability of the model to the organizational systemic level.     

Social class, mental illness, and social mobility: the social selection-drift hypothesis for serious mental illness

The assumptions and methods of previous studies of the social selection-drift hypothesis for serious mental illness are examined by using comtemporary log-linear methods for social mobility analysis. The null hypothesis of no difference in intergenerational social mobility between seriously mentally ill and general population control groups cannot be rejected in previous studies by Birtchnell (1971), Goldberg and Morrison (1963), Langner and Michael (1963), and Turner and Wagenfeld (1967). The findings of this study suggest that previous empirical support for intergenerational social mobility differences is an artifact of not controlling for group differences in origins and destinations when collapsed origin-by-destination tables are analyzed. This study suggests that intergenerational social mobility differences between seriously mentally ill and general population groups in previous studies provide very little, if any, empirical support for social selection-drift processes in serious mental illness.     

Bad Days Don't End When the School Bell Rings: The Lingering Effects of Negative School Events on Children's Mood, Self-esteem, and Perceptions of Parent–Child Interaction

Each day for five days, 79 fifth‐grade children reported on events that occurred at school and they and their parents described their interactions with each other each evening. Consistent with previous research, it was found that on days when children reported more academic or peer problems during the day at school, they later described more aversive interactions with their parents. As hypothesized, increases in anxiety and drops in children's state self‐esteem partially mediated this link. However, parents did not report any differences in their interactions with the target child on days when the child experienced problems at school. This study suggests that negative events experienced by children while at school lead to short‐term changes in mood and self‐esteem, which influence their perceptions of subsequent interactions at home with parents.     

Approach to normality of mean and M-estimators of location

The approach to normality of an estimate is displayed graphically by the nonlinearity in the derivative of the log density, and this nonlinearity is related to the accuracy of the normal approximation for the tail area. Using techniques from small-sample asymptotics, an alternate proof of the central limit theorem is given and same indices of tail behavior are examined.     

Social Policy at the 2004 Election

Social policy appeared to be a key battleground at the 2004 Australian Federal election. Opposition Leader Latham announced major policies on Medicare, family support and taxation, and schools funding during the election campaign. Using sample survey data from the Australian Election Study 2004, this paper analyses how these policies may have influenced voters. In brief, although a significant proportion of electors identified these issues as being extremely important to them when they were deciding about how to vote, many made up their mind about how to vote around the time of the announcement of the election or before. This mitigated the potential effect that these major policies could have on the election outcome. Nevertheless, these policies were important and Labor had a significant policy advantage amongst those who were late deciders about how to cast their vote.     

Infant Auditory Sensitivity to Pure Tones and Frequency‐Modulated Tones

It has been suggested that infants respond preferentially to infant‐directed speech because their auditory sensitivity to sounds with extensive frequency modulation (FM) is better than their sensitivity to less modulated sounds. In this experiment, auditory thresholds for FM tones and for unmodulated, or pure, tones in a background of noise were measured for 4‐month‐old infants using a conditioned response procedure. The FM tones swept from 150 to 275 Hz or from 150 to 550 Hz. The frequency of the pure tone was either 275 or 550 Hz. The results showed that infants were slightly, but significantly, more sensitive to the sounds that included 550 Hz than they were to the lower frequency sounds, whether or not the sound was frequency modulated. It appears that infants could be somewhat more sensitive to infant‐directed than to adult‐directed speech, not because of FM per se, but because the fundamental frequency excursions in infant‐directed speech extend into a higher frequency range than those in adult‐directed speech.