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.     

Children's preference for novelty in selective learning: Developmental stability or change?

Data relating novelty preference to age for normal children are inconsistent, although a current theory predicts a developmental shift from novelty to familiarity preference in selective learning (D. Zeaman, 1976, in T. J. Tighe & R. N. Leaton (Eds.), Habituation: Perspectives from child development, animal behavior, and neurophysiology (pp. 297–320), Hillsdale, NJ: Erlbaum). Support for this theory, however, has been derived primarily from studies of retardate learning. Normal children's novelty preference was examined within a modified Moss-Harlow (E. Moss & H. F. Harlow, 1947, Journal of Comparative and Physiological Psychology, 40, 333–342) design to compare Zeaman's model with that of S. L. Witryol and W. Wanich (1983, The Journal of Genetic Psychology, 143, 3–8). Each of 16 problems, consisting of three single-stimulus demonstration trials and one two-choice test, was administered to 180 children (mean CA 4, 5.5, and 7 years) in three reward conditions. Novelty was manipulated by varying stimulus familiarization in the demonstration trials. Experiment 1 showed strong preferences for novel over familiar (demonstrated) stimuli at each age. Experiment 2 revealed novelty preference across age levels, two levels of reward contrast, and two levels of task difficulty. It was reasoned that Moss-Harlow tasks designed for normal children typically present a much higher level of difficulty than that intended by researchers. Furthermore, developmental decreases in novelty preference by retardates may derive from (a) transfer of training from prior experiments and (b) specific, repetitive instructions which may have directed attention away from stimulus novelty.     

Intelligence and fertility in the United States: 1912-1982

Results are presented for the 1st analysis of the relationship between IQ and completed fertility using a large, representative sample of the US population. Correlations are predominantly negative for cohorts born between 1894 and 1964 but are significantly more positive for cohorts whose fertility was concentrated in the baby boom years. Previous studies reporting slightly positive correlations appear to have been biased in their restriction of samples to atypical cohorts. The National Opinion Research Center (NORC), a nonprofit research organization affiliated with the University of Chicago, conducted the General Social Survey (GSS) in the US each year from 1972 to 1982, except for 1979. A combination of block quota and full probability sampling was employed. Hour-long interviews were completed with 12,120 respondents who were English-speaking, noninstitutionalized adults (18 years or older) living within the continental US. Such questions as age, place of birth, income and occupation, were asked in each interview. Other questions about attitudes on various social, political, and moral issues were rotated in different years. The unique opportunity this data set affords is an overview of the relationship between intelligence and fertility for a nationally representative sample of Americans whose major reproductive years fell between 1912 and 1982. Data were consolidated from the 4 surveys in which the vocabulary test was given (1974, 1976, 1978, and 1982). Respondents were divided into 15 birth cohorts of 5-year intervals ranging from before 1894 to 1964. Correlations between vocabulary scores and number of siblings are markedly negative across all 15 cohorts. Vocabulary sibling correlations are more negative in every cohort than vocabulary offspring correlations. Previous reports of a neutral or slightly eugenic relationship appear to be due to the nature of the samples used, in part because the cohorts chosen were atypical, and in part because they did not include nonwhites. Childless respondents averaged slightly higher scores than did those with 1 or more children, indicating that the automatic exclusion of the childless from sibling-IQ studies has not spuriously inflated negative correlations.     

Group differences in intelligence and speed of information-processing

Three samples of different levels of mental ability, but each of average or above- average intelligence, were administered a battery of reaction time test which measure the speed with which persons can execute various elementary cognitive processes. Discriminant analyses revealed significant differences between the groups in reaction times and intraindividual standard deviations, and univariate analyses indicated that higher ability groups tended to obtain significantly faster mean reaction times and to show significantly less intraindividual variability than lower ability groups. Groups which differed the most in mental ability differed to the greatest extent on the more complex reaction time tests but this pattern was reversed for groups which were relatively close in mean IQ. Generally, the results support the hypothesis that speed- of-processing is an important factor unferlying intelligence, but a number of puzzling findings indicate the need for continued research in this area.     

A measure of agreement for use with the Holland classification system

A measure of agreement for partially ordered data is suggested that makes use of the k categories with the highest ranks, the ranks of the remaining categories being disregarded. The proposed measure is applied to the Self-Directed Search in order to evaluate its agreement with occupational preference. Other forms of agreement in the Holland classification system can also be assessed by the proposed measure. Examples are congruence and husband-wife and counselor-client agreement.     

Midi-level measurement of social anxiety in psychiatric and non-psychiatric samples

This paper reports the development and assessment of midi-level behavioral measures of social anxiety in the context of two experiments, one studying an analog student sample, the other a psychiatric sample. Judgments on nine categories of clinically practical midi-level behaviors (e.g. Facial Expression, Orienting, Sense of Timing), based on a review of the literature on human ethology and on pilot research, were compared to global judgments of social anxiety and social skill and to physiological arousal. Intraclass correlations exceeded 0.80 for judgments of the global and midi-level behavioral ratings on both samples. Results of correlational analyses indicated that while there were several significant predictors of global skill and anxiety among the midis, the magnitude of the relationship between midis and globals was stronger for the patient than the student sample. Further analyses based on S's heart rate (HR) reactivity suggested that while global ratings did not significantly predict H R in a high social anxiety situation, one midi-level behavioral rating (self-manipulations) did. The clinical utility of the newly developed measures is discussed with particular attention to their practicality for behavior therapy.     

Measurement of adult performance on Piaget's water horizontality task

Three versions of Piaget's water-level-horizontality task (multiple choice, drawing, and apparatus tests) were administered to 121 female and 74 male adult volunteers. Sex differences and container orientation effects were replicated. Men were more accurate than women and horizontal and vertical container orientations evoked more accurate responses than oblique orientation. Reliability estimates ranged from .78 to .97 and correlations among the measures ranged from .78 to .88. The relative merits of alternative water-level testing and scoring methods and the orientation and type of stimuli to be used are discussed.     

A unifying model for the structure of intellectual abilities

Models of the structure of cognitive abilities suggested by Spearman, Thurstone, Guilford, Vernon and Cattell-Horn are reviewed. It is noted that some of the models include a general intellectual factor (g) while others do not. It is also noted that some models are nonhierarchical, while in others more narrow abilities are subsumed under broader abilities in a hierarchical pattern. An empirical study in which a test battery of 16 tests was administered to some 1000 subjects in the 6th grade is reported. Using the LISREL technique to test different models, good support is obtained for oblique primary factors in the Thurstone tradition as well as for the second-order factors fluid intelligence, crystallized intelligence, and general visualization hypothesized by Cattell and Horn. It is also found, however, that the second-order factor of fluid intelligence i is identical with a third-order g-factor. On the basis of these results a three-level model (the HILI-model) is suggested, with the g-factor at the top, two broad factors reflecting the ability to deal with verbal and figural information, respectively, at the second-order level, and the primary factors in the Thurstone and Guilford tradition at the lowest level. It is argued that most previously suggested models are special cases of the HILI-model.     

An individualized regression and clustering assessment of the social values of adults and children

Research investigating social values in interdependency situations has generally demonstrated age differences across the 5- to 10-year-old age range. However, the precise nature of these differences is unclear. Further, the differences in the social values of adults and children has not been addressed. The present study examined the social values of undergraduates and children using a measurement technique designed to allow an accurate inference of social values. Ninety-one 6- to 10-year-old children and 100 undergraduates rated the desirability of a set of outcomes which systematically varied with regard to the benefits to themselves and another person. Theoretically specified characteristics of the outcomes were then used to generate a regression equation predicting the desirability ratings separately for each individual. These regression equations were then clustered into groups of individuals who used decision rules characteristic of the major social values. The results indicate that the 8- to 10-year-old children, compared to the 6- to 7-year olds, more often expressed a social value involving equality and tended to less often express a social value involving superiority. Comparison of the social values expressed by the children and adults indicated that the adults more often expressed group-enhancement and individualism social values and less often expressed a social value involving superiority.