Similarities and differences in social attitudes in four Western countries1

A referents scale (single words and short phrases) constructed to measure social attitudes and values was administered to 500 graduate students of education in the United States, 427 mature evening students in Spain, 130 German students, 270 Dutch students, and 685 Dutch adults, a random sample of the Netherlands. The attitudes of the samples were compared by analysis of the means of clusters of referents associated with the factors common to all samples. The means of selected individual referents of social significance were similarly compared across samples. In general, the U.S. student sample and the Netherlands random sample were more conservative than the Spanish, Dutch, and German student samples. On the other hand, the Dutch, Spanish, and American students were more liberal than the Dutch random sample and the German students. The correlations between the profiles of means of referents common to pairs of samples were mostly substantial (average r's of .75 for conservative referents and .62 for liberal referents), indicating considerable cross-cultural agreement in the rank ordering of the referents within the conservative and within the liberal categories, and thus, presumably, common cores of reaction to the two kinds of referents. The use of attitude-value referents, especially in cross-cultural research, was judged valid and useful.     

Sex differences in symbol-digit substitution task performance

Estes (1974) proposed that the superior performance of women over men on the Digit-Symbol Substitution subtest of the Wechsler scales was due to their greater ability to encode symbols verbally. Three forms of a symbol-digit substitution task were administered to 62 female and 96 male college students. The symbol sets of the three forms differed in the relative amount of figural and spatial orientational information. Women performed better than men on the form which used the Wechsler symbol set in which all symbols are figurally different and, presumably, most easily coded verbally. Men performed better than women on the form which had the greatest amount of spatial orientational information. Estes' hypothesis is supported by these findings.     

Estimating major sources of measurement error in individual intelligences scales: Taking our heads out of the sand

Four ubiquitous major sources of measurement error for individual intelligence scales were identified and discussed. It was argued that where these sources cannot be directly investigated, they should be estimated rather than ignored. The typical magnitude of error arising from each of content sampling, time sampling, scoring, and administration was estimated; their composite was found to produce a far larger standard error of measurement than manuals of individual intelligence tests typically report. Implications for practitioners are discussed.     

Modernization and sex roles: The status of women in the Thai bureaucracy

Modernization is conceptualized as a change in socially sanctioned roles. This study investigates the repatterning of sex and occupational roles in the modernizing bureaucracy of Thailand. Although Thai women have traditionally participated actively in family and economic life, such freedom and equality with males has not been carried over into modern occupations. Female Thai civil servants have lower positions and are paid less than their male counterparts with comparable preparation and accomplishments; and the females make greater sacrifices in family roles. Yet women express greater satisfaction with their civil service jobs than men. Explanations for these findings are suggested.     

Effect of perceptual cohesiveness on pattern recoding in the block design task

In the block design task, one type of intelligence test task, the subject reproduces a square-field pattern by assembling cubes having solid and diagonally divided surfaces. The task requires that the subject recode the perceived stimulus pattern into the alternatives available on the block surface. This experiment examined the effect on response or reproduction time of perceptual cohesiveness of pattern, quantified as the number of adjacent same-colored edges of cubes which abutt to form the pattern. This stimulus variable interacts with the type of design (quantified as total set size for designs constructed of all solid surfaces, all diagonally divided surfaces, or mixtures of the two types of surfaces) and with cuing (presence or absence on stimulus pattern of a superimposed grid congruent with block edges of the constructed design). Cohesiveness increases the difficulty of “mentally slicing” or recoding the pattern into the set of block alternatives. The psychological processes involved in recoding probably are the most important ones underlying the intelligence test task.     

Three-membered domains for aristotle's syllogistic

The paper shows that for any invalid polysyllogism there is a procedure for constructing a model with a domain with exactly three members and an interpretation that assigns non-empty, non-universal subsets of the domain to terms such that the model invalidates the polysyllogism.I am grateful to T. J. Smiley and an anonymous referee for comments that improved this paper.     

Rigorous health screening reduces age effect on memory scanning task

Eighty subjects participated in a study with five age groups (20, 30, 40, 50, and 60 years). Forty subjects showed evidence of factors related to brain dysfunction (risk factors). Their performance on a Sternberg-type memory scanning task was assessed. Age-related slowing of virtually all aspects of the memory scanning process was observed in the healthy group. However, the effect of the presence of risk factors was larger than that of biological age. The results of the present study make a reasonable case for the view that many age effects reported in the literature can be largely explained by suboptimal brain functioning, i.e., by other factors than aging per se.