A general model for the analysis of multilevel data

A general model is developed for the analysis of multivariate multilevel data structures. Special cases of the model include repeated measures designs, multiple matrix samples, multilevel latent variable models, multiple time series, and variance and covariance component models.We would like to acknowledge the helpful comments of Ruth Silver. We also wish to thank the referees for helping to clarify the paper. This work was partly carried out with research funds provided by the Economic and Social Research Council (U.K.).     

Self-selection biases in correlational studies based on questionnaires

It is commonly held that even where questionnaire response is poor, correlational studies are affected only by loss of degrees of freedom or precision. We show that this supposition is not true. If the decision to respond is correlated with a substantive variable of interest, then regression or analysis of variance methods based upon the questionnaire results may be adversely affected by self-selection bias. Moreover such bias may arise even where response is 100%. The problem in both cases arises where selection information is passed to the score indirectly via the disturbance or individual effects, rather than entirely via the observable explanatory variables. We suggest tests for the ensuing self-selection bias and possible ways of handling the ensuing problems of inference.The author would like to thank members of the Psychology Department at UWA for references on the illustrative example, namely the effects of sex-stereotyping on achievements in mathematics; thanks are also due to the anonymous reviewers whose comments resulted in material improvements to the paper.     

Facial reactions to auditory stimuli: Sex differences

This study explored whether males and females differ in facial muscle activity when exposed to tone stimuli with different intensity. Males and females were repeatedly exposed to 95 dB and 75 dB 1000 Hz tones while their facial electromyographic (EMG) activity from corrugator and zygomatic muscle regions were measured. Skin conductance responses were also measured. It was found that 95 dB but not 75 dB tones evoked increased corrugator activity. This effect differed significantly between males and females. Thus, it was only females that reacted with a significant increased corrugator response to the high intensity tone. While facial responses differed between the sexes, the skin conductance response patterns did not. Consistent with previous research it is concluded that females are more facially expressive than are males.     

Familial similarity in Type A behaviour and physiological measurements as related to sex

Thirty-one families with pre-school children were examined with regard to Type A behaviour and cardiovascular and neuroendocrine measurements. Type A scores (MYTH) of the children from the ages of three to six were positively correlated with independent Type A measurements (Jenkins Activity Survey, Bortner Type A scale) of the parents (significant for fathers but not for mothers). The children's resting systolic blood pressures were significantly correlated with maternal and paternal values, while no significant familial aggregation was found for diastolic blood pressure, heart rate, catecholamine or cortisol excretion. The children's systolic and diastolic blood pressures and adrenaline excretion were significantly correlated between ages three and six. The results are consistent with the assumption that genetic as well as environmental factors contribute to the development of Type A behaviour. No significant association was found between Type A scores and physiological measurements during routine activities in the children or the adults.     

Patterns of mother-infant interaction and attachment: A cluster-analytic approach

This study examined stability and change in patterns of mother-infant interaction between 5 and 10 months and their relation to attachment security at 18 months. Cluster analysis was used to identify patterns of dyadic interaction. There was significant, but not substantial, stability in cluster membership from 5 to 10 months. For females, being in the most optimal cluster at 5 months related to subsequent attachment security, regardless of the 10-month pattern of interaction. For males, who were significantly more likely to be insecure than females, insecure attachment was predicted by negative change from 5 to 10 months (moving from a more optimal to a less optimal cluster).     

Attentiveness of mothers and fathers to their baby's cries

Thirty-two mother-father-infant triads participated in a study to examine whether parents' responsiveness to their own infant's distress was affected by the infant's birth order and gender. After separating from their 10- to 11-month-old baby, one parent from each family (16 mothers and 16 fathers) was selected to hear cries that were attributed to the child. Consistent with previous findings regarding physiological reactions to crying and observations of caretaking behavior, first-time parents gave quicker and more frequent attention to their infants than did multiporous parents. Speedy intervention was related to the amount of caretaking responsibilities that parents reported they assumed and to the age at which parents believed that their infant first recognized them. The results demonstrate the importance of caretaking for promoting parental sensitivity to infant signals.     

The lntergenerational transmission of family violence: The long-term effects of aggressive behavior

This paper examines the contention that family violence is passed down from generation to generation. The sample was primarily nonclinical and was composed of 14 clients of community guidance clinics and 82 members of a randomly selected control group. The primary hypothesis that individuals who are exposed to a high degree of physical punishment as children are more likely to resort to family violence as adults was supported. Studies of the modeling of aggression were examined to suggest factors which may mediate the hypothesized relationship. It was found that respondents who had grown up in an environment characterized by either low family warmth or high stress in combination with exposure to a high degree of parental punishment were more likely to use physical punishment in their own families. Also, family violence was more likely to be transmitted in same-sex rather than cross-sex linkages.     

Catholic Feminist Ethics Reconsidered

Taking Catholic sexual ethics and liberal feminist ethics as points of departure, this essay argues that both frameworks are ill‐prepared to deal with the moral problems raised by sex trafficking: while Catholic sexual ethics is grounded in a normative understanding of sexuality, liberal feminist ethics argues for women's sexual autonomy, resting upon freedom of action and consent. From a perspective that attends both to the phenomenological interpretation of embodied selves and the Kantian normative interpretation of dignity, it becomes possible to critique both the Catholic and the liberal feminist frameworks of ethics. I argue that Catholic sexual ethics requires a reconceptualization as social ethics in order to meet the challenges of our present time, but that the shift is possible without giving up the moral imperatives of both Catholic and feminist ethics to protect human dignity and women's rights.     

Sex differences in the perceived causal structure of date rape: A preliminary report

The present study examines the perceived causes of date rape by the use of network analysis. The 142 subjects in the study were presented with grids containing 9 causes listed by row and column. They were asked to indicate if they belived there was a causal link between each of the 72 causal pairings. The resulting network placed emphasis on male and female drunkenness, norms of dating, and society's attitude to date rape as distal causes of date rape. Misunderstanding of behavior by men, male attitude that date rape is unimportant, and need by males for dominance were seen as proximal causes. Further networks for each sex were constructed: these showed clear division of opinion on the causes of date rape. © 1995 Wiley-Liss, Inc.     

Effects of neonatal treatment with 1-(2,5-dimethoxy-4-lodophenyl)-2 aminopropane HCI (DOI) and ritanserin on agonistic behavior in adult male and female rats

The role played by the neonatal 5-hydroxytryptamine (5HT) system in the organization and sexual differentiation of adult agonistic behavior was investigated in rats. Focus was on the 5HT2 receptor subtype, which has been demonstrated to be involved in agonism control in the adult. 5HT2 activity was experimentally manipulated by administration of a specific agonist [1-(2, 5-dimethoxy-4-iodophenyl)-2-aminopropane HCl (DOI)] or antagonist (ritanserin) during the second week of life, when serotonin is known to concur to anatomical and behavioral sexual differentiation. Interactions between early 5HT2 activity, genetic sex, and neonatal circulating testosterone (T) were studied by administering the ligands to males, females, and androgenized females. At adulthood, the animals were tested for both aspects of agonism, i. e., aggression and defense, in a 20-min confrontation with an unfamiliar conspecific of the same sex, age, body weight, and social experience. Neonatal administration of the 5HT2 antagonist ritanserin increased aggression independently of sex; it also increased defense, but this effect was confined to males. The agonist DOI had no effect on aggression, but enhanced defense in males and androgenized females, with an effect which depended therefore more on neonatal T than genetic sex. Females appeared in general less sensitive to neonatal 5HT2 manipulation than both androgenized females and males; this suggests that neonatal T is crucial for experimental modifications of neonatal 5HT2 activity to have any consistent effect on adult agonistic behavior. On the other hand, effects observed in males and androgenized females were dependent on the behavior considered and the drug administered. This was especially evident for defense, enhanced by ritanserin in males only, and in both males and androgenized females by DOI. Neonatal 5HT2 activity seems therefore to play a role in the modulation of adult agonistic behaviors, which depends on the behavior considered and is under multiple control of genetic sex and hormonal neonatal substrate. © 1994 Wiley-Liss, Inc.