Penalized optimal scaling for ordinal variables with an application to international classification of functioning core sets

Ordinal data occur frequently in the social sciences. When applying principal component analysis (PCA), however, those data are often treated as numeric, implying linear relationships between the variables at hand; alternatively, non-linear PCA is applied where the obtained quantifications are sometimes hard to interpret. Non-linear PCA for categorical data, also called optimal scoring/scaling, constructs new variables by assigning numerical values to categories such that the proportion of variance in those new variables that is explained by a predefined number of principal components (PCs) is maximized. We propose a penalized version of non-linear PCA for ordinal variables that is a smoothed intermediate between standard PCA on category labels and non-linear PCA as used so far. The new approach is by no means limited to monotonic effects and offers both better interpretability of the non-linear transformation of the category labels and better performance on validation data than unpenalized non-linear PCA and/or standard linear PCA. In particular, an application of penalized optimal scaling to ordinal data as given with the International Classification of Functioning, Disability and Health (ICF) is provided.     

A NOTE ON THE POSSIBLE DETERMINANT OF “SPLIT REPRESENTATION” AS AN ARTISTIC STYLE

L'A. fait L'hypothèse que le dédoublement de la représentation, qui est caractéristique du style graphique de certaines cultures, provient d'une préférence universelle pour ce type de dessin chez les tout jeunes enfants de toutes cultures; mais dans la plupart des cultures, cette préférence serait supprimée parce que, paradoxalement, les dessins doubles ne permettent pas de reconnaître les objets dessinés aussi facilement que les dessins conventionnels qui sont plus proches de la projection rétinienne des objets. Les dessins avec enchaînement ou dédoublement de la représentation seraient conservés chaque fois qu'ils sont appréciés en tant que code utilisé à d'autres fins que L'information sur les objets peints et pour leur valeur esthétique.     

CIRCULAR RELATION BETWEEN UNDERSTANDING AND LOGIC

Procedures for determining logicality presuppose understanding, and procedures for determining understanding presuppose logic. One can escape from this circle only by presupposing logicality, in agreement with common-sense thinking. Understanding can then be studied as an empirical variable. Traditional research has been based on the opposite solution: Logicality is treated as an empirical variable, understanding is implicitly presupposed; hence results are obtained which do not 'make sense'. One cannot understand the illogical.     

DETERMINANTS OF PERFORMANCE ON DOUBLE CLASSIFICATION TASKS

Two groups of children, median ages 6–7 and 8–7, were presented with objects examplifying three combinations of two dichotomous attributes, and asked to indicate the remaining combination. Keeping objects covered vs. uncovered at the time of the test response had generally little effect. Labelling was consistently superior to perceptual matching, apparently because the latter involved a search process. The errors in the labelling condition consisted mostly in naming an object identical to or different in both attributes from those given, reflecting perhaps an inability to break down the impression of the objects into independent attributes.     

Prevalence of personality disorders among bulimics,nonbulimic binge eaters,and normal controls

We examined the comorbidity of bulimia and personality pathology among college women. Subjects included women (n=23) meeting DSM-III-R criteria for bulimia nervosa (bulimics), women (n=23) who reported binge eating but did not fulfill the criteria for bulimia nervosa (binge eaters), and women (n=23) who did not binge eat (normals). The subjects completed an assessment battery consisting of the Personality Disorders Examination and the SCID as well as the Beck Depression Inventory, the Rosenberg Self-Esteem Index, and measures of impulsivity and self-defeating tendencies. Fourteen of 23 bulimics (61%) met criterion for a personality disorder using DSM-III-R Axis II criteria. In comparison, 3 of 23 (13%) binge eaters and 1 of 23 (4%) normal subjects received an Axis II diagnosis. Borderline and self-defeating diagnoses were the predominant personality disorders in 96% of the bulimics exhibiting clinically significant personality pathology. Bulimics also exhibited significantly more depression, impulsivity, and self-defeating behavior and lower self-esteem than binge eaters and normals. The findings are discussed within a conceptual framework that posits an interaction between personality pathology and restrained eating.     

Shape from stereo: A systematic approach using quadratic surfaces

We used quadratic shapes in several psychophysical shape-from-stereo tasks. The shapes were elegantly represented in a 2-D parameter space by the scale-independentshape index and the scale-dependentcurvedness. Using random-dot stereograms to depict the surfaces, we found that the shape of hyperbolic surfaces is slightly more difficult to recognize than the shape of elliptic surfaces. We found that curvedness (and indirectly, scale) has little or no influence on shape recognition.     

On becoming a leader: Effects of gender and cultural differences on power distance reduction

The present research focused on power processes in a simulated organizational structure consisting of three hierarchical levels occupied by different numbers of mules and females. Subjects were presented with a chart showing the organizational hierarchy of which they were a member placed at the lowest level, and asked to nominate any person for the leader position vacated by the current incumbent. The results of Experiment I (n = 88 Dutch male and female university students) showed that male subjects strongly overnominated themselves, whereas a majority of the female subjects nominated either self or another female. Of the others that were nominated by both males and females, almost all were occupants of positions immediately below the leader position, indicating the normative influence of a bureaucratic rule of leader succession. Experiment 2 was a replicational study carried out in a different culture (n = lOl Polish male and female university students). Polish subjects adhered to the bureaucratic rule more strongly than their Dutch counterparts, and both females and males nominated mostly males. Results are discussed with reference to gender self-stereotypes and cultural differences.     

Attitude change following persuasive communication: Integrating Social Judgment Theory and the Elaboration Likelihood Model

An experiment was conducted to examine the influence of the perceived extremity of a message and motivation to elaborate upon the process of persuasion. The first goal was to test a model of attitude change relating Social Judgment Theory to the Elaboration Likelihood Model. The second objective was to develop an instrument to measure attitude structure (latitudes of acceptance, non-commitment, and rejection) that allowed for a more refined assessment of the discrepancy between the position advocated in a message and the recipient's initial attitude. The main dependent variable was the attitude towards the use of automobiles in relation to environmental issues. Subjects were confronted with a message located in their own latitude of acceptance, rejection or non-commitment. Shortly after, a second measurement of attitude took place. The results showed that messages within the latitudes of non-commitment gave rise to the greatest attitude change. The data support the susceptibility hypothesis that subjects elaborate messages mainly in the latitude of non-commitment.