Examining distributional characteristics of clusters.

Standard cluster analysis creates clusters based on the criterion that their members be closer to each other than to members of other clusters. In this article, it is proposed to examine empirical clusters that result from standard clustering, with the goal of assessing whether they contradict distributional assumptions. Four models are proposed. The models consider two data generation processes, the Poisson and the multinormal, as well as two convex shapes of cluster hulls, the spherical and the ellipsoidal. Based on the model, the probability of being in a cluster of a given location, size, and shape is estimated. This probability is compared with the observed proportion of cases. The observed proportion can turn out to be larger, as large, or smaller than expected. Examples are given using simulated and empirical data. The simulation showed that the size of a cluster, the data generation process, and the true distribution of data have the strongest effect on the results obtained with the proposed method. The empirical examples discuss distributional characteristics of cross-sectional and longitudinal clusters of aggressive behavior in adolescents. The examples show that clustering methods do not always yield clusters that contradict distributional assumptions. Some clusters contain even fewer cases than expected.

Women survivors of intimate partner violence and post-traumatic stress disorder: prediction and prevention.

A considerable body of research has demonstrated that women who are abused by their male romantic partners are at substantially elevated risk for the development of post-traumatic stress disorder (PTSD). This article reviews recent literature regarding intimate partner violence (IPV) and resultant PTSD symptoms. The article is intended to be an introduction to the topic rather than an exhaustive review of the extensive literature in this area. Factors that enhance and reduce the risk for PTSD, including social support, coping styles, and types of abusive behavior experienced, are described. In addition, the unique risks associated with IPV for women who have children are discussed. Prevention efforts and treatment are briefly reviewed.

Functional configuraI frequency analysis: explaining types and antitypes.

In this article, we ask how types and antitypes in Configural Frequency Analysis (CFA) can be explained at the level of effects in log-linear models. Two strategies are proposed for the search for the effects that cause types and antitypes to emerge. The first strategy is ascending and inclusive. Beginning with the CFA base model, effects are systematically added until the most parsimonious model is found that meets the following two conditions. The first condition is that all types and antitypes disappear. The second condition is that no new types and antitypes emerge. The second strategy is descending and exclusive. Beginning with the saturated model, effects are systematically excluded until (1) the effects are identified that would make all types and antitypes disappear, and (2) no new types and antitypes emerge. A first comparison of the two models suggests that the ascending, inclusive strategy yields more parsimonious models. However, results from the descending, exclusive strategy have a more clear-cut interpretation. Data examples are given that describe the development of severe domestic violence.

Parental alcoholism and co-occurring antisocial behavior: prospective relationships to externalizing behavior problems in their young sons.

The hypothesis that parental alcoholism and co-occurring antisocial behavior would be indirectly linked to child externalizing behavior problems through child lack of control, current levels of parent depression, family conflict, and parent-child conflict was tested using manifest variable regression analysis. Participants were a community sample of 125 families with an alcoholic father and 83 ecologically matched but nonsubstance abusing families involved in the first 2 waves of an ongoing longitudinal study (with 3 years between each wave). All families had a biological son who was 3-5 years old at study onset. Results revealed that child lack of control mediated the relation between paternal alcoholism and the son's subsequent externalizing behavior problems. Family conflict was a significant mediator of maternal and paternal lifetime antisocial behavior effects and father-son conflict mediated paternal lifetime antisocial behavior effects. Study implications are discussed within the context of parental socialization of antisocial behavior.

The odds of resilience.

This article presents sample research designs for the investigation of questions concerning resilience. Sample hypothesis from specific research designs are described in the form of odds ratios.

Text and graphics: manipulating nutrition brochures to maximize recall.

This study examined how altering text and graphics of a nutrition brochure could affect the ability to remember the content of the message. Two theoretical models were used to guide alterations: dual-coding theory and the communications model. Three brochure formats were tested: the original brochure containing abstract text and abstract graphics, a modified brochure with relatively concrete text and abstract graphics, and a relatively concrete text brochure with concrete graphics. Participants (N = 239 women) were divided into four age groups: 20-30, 40-50, 60-70 and over 70 years. Women were randomly assigned into each of the three experimental brochure formats or a control group. Participants completed recalled materials from the assigned brochures (the no treatment control group did not include a brochure) at two different sessions, 30 days apart. Data were content analyzed and results were compared using analysis of covariance to test differences by age and brochure types. Younger women (20-30 and 40-50 years) recalled more information than women over 60 years. More concrete nutrition education print materials enhanced recall of information presented immediately after reading the material; however, this effect was transient and lasted less than 30 days after a one-time reading. The implications of these data for communicating nutrition messages with print materials are discussed.

Does self-competence predict gender differences in adolescent depression and anxiety?

This longitudinal study examined 75 young adolescents to explore whether self-competence predicts the emergence of gender differences in depression and anxiety. During both 6th and 7th grade, boys reported significantly higher levels of self-competence than did girls. In addition, boys were significantly less depressed and anxious than girls in 7th grade, but not in 6th grade. Finally, when the variance contributed by self-competence was accounted for, the relationship between gender and trait anxiety weakened and the relationship between gender and depression became non-significant. These results support the hypothesis that self-competence is partially responsible for the emergence of gender differences in depression and anxiety during early adolescence.

Perceived parental acceptance and early adolescent self-competence.

Relationships between perceived parental acceptance and adolescent self-competence were examined in 214 sixth- and seventh-grade students by both adolescent and parental gender. Results indicated that for boys, paternal but not maternal, acceptance significantly predicted self-competence, while the opposite pattern was found for girls. In addition, self-worth significantly predicted maternal and paternal acceptance for both boys and girls. The consistency of these findings with current models of human development is discussed, along with their implications for research.

The symmetry of symptom patterns in pre-post treatment designs.

This article proposes a new nonparametric method for statistical evaluation of clinical pre-post treatment designs. In clinical research, models of marginal symmetry typically are estimated from log-linear models of axial and quasi-symmetry. As such, they provide overall goodness-of-fit information concerning change in probabilities of categories of one variable that was observed twice. This paper proposes the following three extensions: (1) using models of marginal symmetry for changes in patterns of two or more variables, and (2) following up global marginal symmetry tests using Lehmacher's sign tests. (3) To protect the experiment-wise alpha, a modified Bonferroni-Holm procedure is proposed. The new approach allows researchers to make statements about treatment effects at the level of single symptoms. Examples illustrate application of all three symmetry models and the follow-up test using data from pharmaco-psychiatry. The discussion relates Lehmacher's tests to two-sample Configural Frequency Analysis of multi-discrimination types. Strategies of statistical significance testing are presented and the importance of the proposed methodological approach for psychiatric research is discussed.

Use of qualitative and quantitative methods to define behavioral fat-reduction strategies and their relationship to dietary fat reduction in the Patterns of Dietary Change Study.

OBJECTIVE: The purposes of this study were to identify specific food choice behaviors used to decrease dietary fat intake in a community-dwelling population; assess how people categorize changes to diet; determine whether logical grouping of food intake changes revealed one or more common patterns or strategies used by these participants to decrease fat; and determine which strategies were responsible for the greatest decrease in dietary fat intake in the study population. DESIGN: Survey analysis and in-depth interviews were used to quantitatively and qualitatively define dietary change patterns retrospectively in a population who, according to self-report, had decreased their fat intake. Specific food changes made to decrease fat intake, interview statements, and participants' reduction of percentage energy from fat were examined. SETTING: Interviews were conducted from June 1993 through April 1994. SUBJECTS: Included in the study were 145 persons aged 30 to 55 years who reported that they had been decreasing their dietary fat intake for 5 years or more, maintained a healthful diet for at least 5 months, and resided in the United States while changing their diets. STATISTICAL ANALYSES PERFORMED: Confirmatory factor analysis, reliability analysis, and linear regression analysis were performed. RESULTS: Nine fat-reduction strategies were identified. Decrease fat flavorings, decrease "recreational foods," decrease cooking fat, replace meat, change breakfast, and use fat-modified foods accounted for significant reduction in fat intake. CONCLUSIONS: People use a variety of dietary changes to reduce their fat intake. These changes can be categorized into strategies according to the way people change their diets. Knowledge of these strategies and their importance in dietary fat reduction can improve and help nutritionists prioritize the messages they convey.

Response to oCRH in depressed and nondepressed adolescents: does gender make a difference?

OBJECTIVE: To examine the hypothesis that hypothalamic-pituitary-adrenal responses to stress vary across gender, contributing to gender differences in the prevalence of depression. METHOD: This study examined gender differences between depressed (n = 21) and control (n = 20) adolescents in adrenocorticotropic hormone (ACTH) and cortisol response to two ovine corticotropin-releasing hormone (oCRH) tests, at baseline and following a cognitive stressor. RESULTS: Boys had higher (p < .05) measures of ACTH than girls, regardless of depression status, whereas corresponding cortisol parameters were similar in both groups. Cortisol measures were higher (p < .05) at time 1 than at time 2 in both groups, a phenomenon that might reflect the novelty of the situation. CONCLUSIONS: Gender differences in hormone responses may be related to differences in peripheral metabolism of ACTH, resulting in changes of immunoreactivity but not bioactivity or a different set point of the hypothalamic-pituitary-adrenal axis. The pattern of ACTH and cortisol responses to oCRH and the 24-hour excretion of free cortisol was normal in adolescents with depression, probably reflecting normal negative feedback mechanisms at this age or that most of these patients suffer from atypical rather than melancholic depression.

The aging semantic differential: a confirmatory factor analysis.

The Aging Semantic Differential (ASD), which is used to measure attitudes and quantify bias and negative stereotypes toward older people, is a 32-item scale published more than 20 years ago (Rosencranz & McNevin, 1969). Several factor analytic studies failed to replicate the original three-factor structure. Confirmatory factor analyses (CFA) using 100 third-year medical students compared six-factor models derived from the gerontological literature. A modified version of Holtzman, Beck, and Kerber's (1979) four-factor model proved to be the "best" comparative fit based upon a consensus of fit indices.

Assessment of neural network models using prediction analysis.

Examination of classification or confusion matrices in neural network models is often advocated as a way to investigate the effects of network architecture on classification rules and to understand the topography of the classification space. In addition, for model comparison purposes it is useful to be able to make statements concerning the statistical reliability of these patterns of hit and error cells. Prediction analysis provides a number of descriptive and inferential measures for the evaluation of hypotheses about patterns of hit and error cells. The present paper applied recent advances in prediction analysis to the examination of neural network classification matrices. Results of the application of prediction analysis to the generalized XOR problem and to data from classification of cardiovascular responses of human subjects were used to illustrate the determination of absolute and relative fit of omnibus models as well as the evaluation of specific hypotheses within these models.

The relationship between aggressive behavior and puberty in normal adolescents: a longitudinal study.

PURPOSE: To assess changes in aggressive behaviors as related to progression from early to late puberty in normal adolescents. METHODS: Subjects were normal English schoolchildren. An observational cohort design was employed. Pubertal status was measured by Tanner staging. Self-reports of verbal aggression against adults, physical aggression against peers, aggressive impulses and aggressive inhibitory responses, were collected at three points in time. Analyses were performed for the entire group of 106 subjects in 1983, 77 subjects in 1985 and 70 subjects in 1987. Statistical methods included analysis of variance, regression and correlation and cluster formation. RESULTS: There were decreases in all aggression variables except in aggressive impulses over this time period. When analyzed by gender, boys were initially more aggressive than girls, but by late puberty all gender differences in self-reported aggressive behaviors had disappeared. When only those subjects who were evaluated at all three data collection times were grouped by similar responses on both aggression and physical variables, three clusters of boys and girls were identified. Clusters contained varying proportions of boys and girls. Cluster one (48.5% of the entire sample) was a low aggression group. Cluster two (30.3%) was a high aggression group, and cluster three (21.2%) was an intermediate aggression group. These clusters seemed to have relatively stable aggression characteristics over time. CONCLUSIONS: These data suggest that groups of boys and girls who report similar aggression characteristics and have similar growth and pubertal characteristics can be identified. Neither gender alone, nor pubertal status alone, nor by inference, hormones alone is sufficient to explain the complex set of behaviors which are involved in aggression.

Temperament and adjustment in young adulthood: a 15-year longitudinal analysis.

The relationship between easy/difficult temperament and adjustment from middle adolescence to young adulthood was examined. Cluster analysis revealed four groups of individuals via composite temperament scores over three points of measurement. Multivariate analyses indicated that extremely difficult temperament was associated with poorer psychosocial functioning in childhood and young adulthood.

[Configuration frequency analysis: Kieser's test against Victor's type and syndrome]. / Die Konfigurationsfrequenzanalyse: Kiesers Test gegen Victors Typen und Syndrome.

Tests against types and anti-types as performed in Victor's Configural Frequency Analysis (CFA) have shown to be non-conservative. The conservative test by Kieser is introduced. It is applied to data describing a sample that performs below average in language tests. For data including small expected cell frequencies the Harms-Ihm adjustment is recommended. It leads to valid chi-squared testing even if expectancies are small. The problem of identifying more than two types or syndromes in exploratory CFA is discussed.

Types of change: application of configural frequency analysis in repeated measurement designs.

Many researchers are concerned both with intraindividual change patterns and interindividual differences and similarities in those change patterns. Configural Frequency Analysis (CFA) provides a way to identify overrepresentations (types) and underrepresentations (antitypes) in the frequencies of multiple variable classifications organized to reflect patterns of change. Three methods of CFA for analyzing repeated measures data are considered. To establish trends, two of them require at least ordinal data and the third requires interval data. Data analysis and the interpretation of results are illustrated. CFA is compared with residual analysis from log-linear modeling.

Efficacy of interventions with caregivers: a reanalysis.

This paper presents a reanalysis of data from a previously reported study with family caregivers of dementia patients using the method of prediction analysis. Compared with subjects on a waiting list or enrolled in support groups, caregivers in individual and family counseling were more likely to have successful outcomes on all dependent measures (Brief Symptom Inventory, personal strain, and role strain).

[Syndrome combinations as meta-syndromes in configuration frequency analysis]. / Syndromkombinationen als Metasyndrome in der KFA.

The paper defines metasyndromes as syndrome combinations. Within the framework of Configural Frequency Analysis the authors propose using a single test for identifying patterns of syndromes rather than applying simultaneous tests for identifying singular syndromes. The authors describe and illustrate the method using the "classical data" that describe symptom patterns under LSD (Lienert, 1971). The rationale of the new procedure is to combine independent tests as was proposed by Wartmann and Wette (1952) as a means for meta analysis.