Imputation of missing categorical data by maximizing internal consistency

This paper suggests a method to supplant missing categorical data by reasonable replacements. These replacements will maximize the consistency of the completed data as measured by Guttman's squared correlation ratio. The text outlines a solution of the optimization problem, describes relationships with the relevant psychometric theory, and studies some properties of the method in detail. The main result is that the average correlation should be at least 0.50 before the method becomes practical. At that point, the technique gives reasonable results up to 10–15% missing data.We thank Anneke Bloemhoff of NIPG-TNO for compiling and making the Dutch Life Style Survey data available to use, and Chantal Houée and Thérèse Bardaine, IUT, Vannes, France, exchange students under the COMETT program of the EC, for computational assistance. We also thank Donald Rubin, the Editors and several anonymous reviewers for constructive suggestions.     

The development of general competence and domain-specific competencies

The development of competence was studied in a longitudinal sample of 98 children, with measurement points when the children were 7, 10, and 12 years old. Competence was defined in terms of adequate adaptation to developmental tasks, both on a general and on a domain-specijic level. The general form of competence was measured by ego-resiliency, reflecting fexibility and ability to solve developmental problems Domain-specific competencies were measured by school achievement and social preference. The results showed that ego-resiliency was related to the two domain-specific competencies, although the hypothesized causal direction from general competence to specific competencies was not supported. Both domain-specific competencies were found to have a feedback effect on ego-resiliency. This feedback effect was stronger for social preference. These results suggest a transactional relationship between general and domain-specific competencies, and point to the importance of domain-specific competencies (especially social preference by the peer group) in the development of general competence in the child.     

Confidence in expectations: A test of the triangle hypothesis

Kelley and Stahelski's (1970) triangle hypothesis states that competitors hold homogeneous views of others by assuming that most others are competitive, whereas cooperators or pro-social people hold more heterogeneous views by assuming that others are either cooperative or competitive. To evaluate the triangle hypothesis, this study examines differences between pro-socials, individualists, and competitors not only in their expectations about others' choice behaviour, but also in the confidence with which such expectations are held. It was found that pro-social subjects expected more cooperation than individualists and competitors. More importantly, as predicted on the basis of the triangle hypothesis, pro-socials were less confident about their expectations than competitors, with individualists holding intermediate levels of confidence.     

The Additive Factor Method: A differential diagnostic tool in hyperactivity and learning disability

The Additive Factor Method assumes that task performance is the sum of sequential and independent processes. We studied the duration of the central processes (memory search and decision) and the motor decision process in hyperactive and learning-disabled children under socalled divided attention and S-R compatibility conditions. It was found that the learning-disabled were impaired in memory search and decision processes whereas hyperactives were impaired in the motor decision process.The authors wish to thank L. Leertouwer for making the drawings.     

Effects of gaze manipulation on subjective evaluation of neutral and phobia-relevant stimuli. A comment on Drake's (1987) 'Effects of Gaze Manipulation on Aesthetic Judgments: Hemisphere Priming of Affect'

Using neutral stimuli (mushrooms and flowers) and phobia-relevant stimuli (snakes and spiders), the present investigation examined the effect of gaze shifts on subjective evaluation of stimuli in female subjects (N = 29). One group of subjects had to orient to the right, whereas the other group had to orient to the left in order to see the stimuli. When left-handed subjects were excluded from the analysis, the results showed that subjects in the 'orient right' group gave a significantly more positive evaluation than subjects in the 'orient left' group. The results provide a replication of the phenomenon described by Drake (1987).     

Restrained eaters are rapidly habituating sensation seekers

Several authors have pointed to similarities between eating disorders and addictive behaviors. In earlier studies, addicts were found to score high on the Sensation Seeking Scale (SSS) and to habituate rapidly to neutral stimuli. In this study, we found experimental support for an addiction model of eating disorders: restrained eaters also scored significantly higher on the Sensation Seeking Scale and also habituated significantly quicker to a series of neutral stimuli than unrestrained eaters. No clear evidence was found for the hypothesis that restrained eaters score lower on measurements of anxiety. It is hypothesized that rapid habituation promotes sensation seeking, which may be manifested in excessive consumption of either drugs or food. Restraint may be a strategy to prevent negative consequences of sensation seeking and excessive consumption. The model is related to earlier experimental findings.     

Energy Conservation: Two Easy Ways Out1

This study investigated attitude-behavior relationships in the context of energy conservation. Results indicated that general environmental concern is a poor predictor of energy-conservation behaviors. Direct evaluative comparisions of various non-conservationist behaviors with a more conservationist alternative showed a clear relationship with behavioral preference. Non-conservationists, however, evaluated their own behavior only marginally favorably and in some cases clearly unfavorably. This was accompanied by an overestimation of the common occurrence of these behaviors among the general population and by an unwillingness to relate these behaviors to personality characteristics. It is argued that these biases (i.e., considering one's behavior as a habit that is shared by many others) could hinder behavioral change. These findings are discussed in terms of effective public policy on the issue of energy consumption.     

The effects of social motives on behavior in social dilemmas in two cultures

Subjects in Santa Barbara, California, and Groningen, The Netherlands, participated in a seven-person social dilemma game, presented in terms of a conservation of resources problem. Prior to their decision making in the social dilemma game, subject's social motive (altruistic, cooperative, individualistic, competitive) was assessed by means of two different classification procedures. On the basis of previous research findings American subjects were expected to display relatively more competitive social motives, and Dutch subjects relatively more cooperative ones. However, no indications of crosscultural differences were found neither with regard to the distribution of social motives nor with regard to the amount of resources taken for self in the social dilemma game. In both locations, competitive subjects took most resources for self, individualistic subjects took next most, cooperative subjects took less than individualists, and altruistic subjects took the least. In addition to predictive validity, indices of the convergent validity of two social motive assessment procedures were described.     

Correspondence analysis used complementary to loglinear analysis

Loglinear analysis and correspondence analysis provide us with two different methods for the decomposition of contingency tables. In this paper we will show that there are cases in which these two techniques can be used complementary to each other. More specifically, we will show that often correspondence analysis can be viewed as providing a decomposition of the difference between two matrices, each following a specific loglinear model. Therefore, in these cases the correspondence analysis solution can be interpreted in terms of the difference between these loglinear models. A generalization of correspondence analysis, recently proposed by Escofier, will also be discussed. With this decomposition, which includes classical correspondence analysis as a special case, it is possible to use correspondence analysis complementary to loglinear analysis in more instances than those described for classical correspondence analysis. In this context correspondence analysis is used for the decomposition of the residuals of specific restricted loglinear models.     

Attributional self-presentation and information available to the audience

Public ability attributions of public task performance were investigated as a function of test performance (high or low), task performance (high or low), and the availability of information about test performance to the audience (audience informed or audience not informed). The results were largely in agreement with self-presentation expectations, Ability attributions concerning public task performance addressed to an audience which was not informed about test results, were found to be strongly self-serving, although they were also influenced by consistency of test and task performance. In ability attributions directed at an informed audience, self-serving biases were completely absent, the attributions being in accordance with the consistency or inconsistency of test and task performances.