Situation Taxonomy and Behavioral Diagnosis Using Prospective Self-Monitoring Data: Application to Dietary Adherence in Patients with Type 1 Diabetes

The goals of this study were to create a taxonomy of problem situations for diet adherence in persons with insulin-dependent diabetes mellitus (Type 1 diabetes) and a behavioral diagnosis by grouping patients together who have similar adherence problems. In study one, 29 patients with Type 1 diabetes kept food diaries for 4 weeks. Adherence was assessed using eight behavioral measures: (1) compliance with a meal plan, (2) energy intake, (3) protein, (4) carbohydrates, (5) fat, (6) self-reported overeating, (7) self-reported undereating, and (8) ratings of impulsive eating. The occurrence of each behavior was quantified in a wide range of situational contexts including meal, social context, physical location, and mood. Cluster analysis of behaviors, situations, and behaviors-in-situations was used to create a hierarchical classification model consisting of five major categories of adherence problems: (1) compliance with meal plan, (2) undereating, (3) overeating, (4) impulsive eating, and (5) lifestyle differences. Within each category, situations were associated with everyday eating habits and special circumstances such as social events, and meals in restaurants. Grouping patients based on the similarity of their adherence problems resulted in five diagnostic groups: (1) emotional binge eaters, (2) restrained eaters, (3) unrestrained eaters, (4) undereaters, and (5) cyclical diet bingers. In study two, 28 subjects kept diaries during an intervention program designed to enhance dietary adherence. Study two results were similar to study one. Situation taxonomy and behavioral diagnosis are potentially useful ways to understand the intervention needs of those attempting maintaining dietary behavior changes.     

Parenting typologies in Georgian and Belgian adolescents: Associations with self-esteem

The parenting context is of fundamental importance for the optimal development of adolescents. Making use of a person-centered approach, we derived parenting typologies from the perspective of Georgian and Belgian adolescents based on four dimensions of perceived parenting (responsiveness, structure, psychological control, and autonomy support). Further, we examined how perceived parenting typologies were associated with adolescents' self-esteem. The study sample included 511 Georgian and 830 Belgian adolescents (N = 1341; 48% female). A cluster-analytic procedure on the full sample was conducted separately for mothers and fathers. The analysis yielded five parenting typologies: Supportive—Highly structuring (high responsiveness and structure), Highly structuring—Moderately controlling (high structure and psychological control, moderate—responsiveness), Supportive—Low structuring/controlling (high responsiveness and autonomy support, low structure and psychological control), Highly controlling (very high psychological control, low responsiveness and autonomy support) and Uninvolved (low on all dimensions). Results indicated that the Highly structuring—Moderately controlling paternal cluster was more prevalent in the Belgian sample. A country-moderating effect was found with the paternal Uninvolved profile associated with low levels of self-esteem in Belgian adolescents, but with moderate levels of self-esteem in Georgian adolescents.     

Fuzzy partition models for fitting a set of partitions

Methodology is described for fitting a fuzzy consensus partition to a set of partitions of the same set of objects. Three models defining median partitions are described: two of them are obtained from a least-squares fit of a set of membership functions, and the third (proposed by Pittau and Vichi) is acquired from a least-squares fit of a set of joint membership functions. The models are illustrated by application to both a set of hard partitions and a set of fuzzy partitions and comparisons are made between them and an alternative approach to obtaining a consensus fuzzy partition proposed by Sato and Sato; a discussion is given of some interesting differences in the results.We are grateful to Dr. M.G. Pittau for carrying out the analyses of the macroeconomic data using the method of Sato and Sato (1994).     

A Comparison of Heuristic Procedures for Minimum Within-Cluster Sums of Squares Partitioning

Perhaps the most common criterion for partitioning a data set is the minimization of the within-cluster sums of squared deviation from cluster centroids. Although optimal solution procedures for within-cluster sums of squares (WCSS) partitioning are computationally feasible for small data sets, heuristic procedures are required for most practical applications in the behavioral sciences. We compared the performances of nine prominent heuristic procedures for WCSS partitioning across 324 simulated data sets representative of a broad spectrum of test conditions. Performance comparisons focused on both percentage deviation from the “best-found” WCSS values, as well as recovery of true cluster structure. A real-coded genetic algorithm and variable neighborhood search heuristic were the most effective methods; however, a straightforward two-stage heuristic algorithm, HK-means, also yielded exceptional performance. A follow-up experiment using 13 empirical data sets from the clustering literature generally supported the results of the experiment using simulated data. Our findings have important implications for behavioral science researchers, whose theoretical conclusions could be adversely affected by poor algorithmic performances.     

Observation of Cu nanometre scale clusters formed in Fe85Si2B8P4Cu1 nanocrystalline soft magnetic alloy by a spherical aberration-corrected TEM/STEM

Microstructure of a nanocrystalline soft magnetic Fe₈₅Si₂B₈P₄Cu₁ alloy (NANOMET^®) was investigated by the state of the art spherical aberration-corrected TEM/STEM. Observation by TEM shows that the microstructure of NANOMET^® heat treated at 738 K for 600 s which exhibits the optimum soft magnetic properties has homogeneously distributed bcc-Fe nanocrystallites with the average grain size of 30 nm embedded in an amorphous matrix. Elemental mappings indicate that P is excluded from bcc-Fe grains and enriched outside the grains, which causes to retard the grain growth of bcc-Fe crystallites. The aberration-corrected STEM-EDS analysis with the ultrafine electron probe successfully proved that Cu atoms form nanometre scale clusters inside and/or outside the bcc-Fe nanocrystallites.     

Cluster Analysis Examination of Serial Killer Profiling Categories: A Bottom‐Up Approach

Two studies explored the validity of dichotomous classification of organised/disorganised serial killers and the four typologies (visionary, mission, hedonistic, and power/control) adopted by the Federal Bureau of Investigation (FBI). Tables documenting crime scene criteria were devised consisting of 50 typifying different crime scenes (Study 1) and 48 crime scenes (taken from the 50 crime scene criteria) with a further 10 motive‐based crime scene criteria (Study 2). Adopting content analysis, crime scenes depicted for 40 (Study 1) and 40 (Study 2) serial killers using secondary sources of data were dichotomously coded for the presence or absence of the crime scene criteria. These data were inputted for agglomerative hierarchical cluster using Ward's method as the clustering algorithm. Differences in the nature of clusters were found for male and female serial killers; however, there was no empirical support for the organised/disorganised classifications or the four typologies. The ‘bottom‐up’ approach resulted in clusters from the different crime scene criteria that helped to understand how these criteria were associated within the serial killer cohort considered. It is concluded that the resulting clusters in both studies support the notion of there being an underlying organised element to most serial murder and that serial murders differ according to the nature of disorganised crime scene criteria present. Copyright © 2011 John Wiley & Sons, Ltd.     

Orientational high-entropy alloys

In high-entropy alloys (HEA), the configurational entropy arising from the presence of multiple elements, stabilizes a disordered solid solution in preference to the possible formation of compounds. In the current work, we identify cluster compounds (of the type AM₄X₈) as orientational analogues of HEA (as a first report on orientational high-entropy systems). In cluster compounds, orientational disorder increases the entropy and plays a role analogous to positional disorder in HEA. In the GaMo₄S₈ compound, at temperatures greater than 50 K, the entropic benefit more than makes up for the strain energy cost and stabilizes the disordered phase in preference to an orientationally ordered compound.     

Linking Youth Internet and Conventional Problems

ABSTRACT

This article utilizes data from a clinical sample of 512 youth to examine whether various problematic Internet experiences are distinctly different from or extensions of the conventional adolescent mental and behavioral health problems seen by clinicians. A Two-step Cluster Analysis identified four mutually exclusive groups of youth, those with: (a) online victimization; (b) inappropriate sexual behavior online; (c) online isolation; and (d) online and offline problems. Results suggest support for the idea that problematic Internet experiences are often extensions of experiences and behaviors that clinicians were working with prior to the advent of the Internet. However, the Internet may be introducing something qualitatively or quantitatively new, such as an increased severity, an increased frequency, or some unique dynamic that requires new responses or interventions.     

Sexual Harassment: Prevalence,Outcomes, and Gender Differences Assessed by Three Different Estimation Methods

This study investigates gender differences in prevalence and effects of sexual harassment using three different estimation methods. In a representative sample of 2,349 Norwegian employees, 1.1% self-labeled as targets of sexual harassment, whereas 18.4% reported exposure to sexually harassing behaviors during the last six months. When employing latent class cluster analysis as a method for estimating prevalence of sexual harassment, 2.2% could be classified as targets of frequent harassment and 19.1% could be classified as targets of unwanted sexual attention. Although more women than men self-labeled as targets of sexual harassment, men reported the same number of sexually harassing behaviors as women. Sexual harassment was found to be significantly related to mental health problems and low job satisfaction among men and women.     

Types of self-reported anxiety in outpatients with DSM-III-R anxiety disorders1

To ascertain whether psychiatric outpatients can be classified into distinct types according to their self-reported symptoms of anxiety, the Beck Anxiety Inventory (BAI) was administered to 655 outpatients diagnosed with DSM-III-R anxiety disorders. Cluster analysis identified three internally consistent subscales representing subjective, somatic, and panic symptoms. Further analysis revealed six types of outpatients reflecting below average, panic-subjective, low subjective, low somatic-panic, above average, and subjective-somatic anxiety. The types were differentiated with respect to age, primary diagnosis, clinically rated anxiety, and both self-reported and clinically rated depression.