Patterns of early cognitive development among boys in four subcultural groups.

600 boys divided by age (6, 7, and 8 yrs), race (Black and White), and social class (middle and lower) were given tests of memory (free recall and digit span), intelligence (Peabody Picture Vocabulary Test and Raven Coloured Progressive Matrices), learning (paired-associates and a matrices training task), and transfer (paired-associates and pre- and postmatrices test). There were (a) significant age advances for all groups on all measures, (b) social class differences on learning and intelligence tests, and (c) racial differences on intelligence, digit span, and paired-associate learning. All groups transferred in a similar manner. Factor patterns across age included more similarities (including similar developmental changes) for the different groups than were expected. (29 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Relations between maternal reinforcement and help and the achievement motive in normal-hearing and hearing-impaired sons.

Correlated the observed behaviors of 64 mothers helping their sons perform 2 verbal and 2 nonverbal tasks with the autonomous and social comparison achievement motive scores for 31 hearing-impaired boys and 33 normal-hearing boys, aged 8-12 yrs. Autonomous achievement motive scores were positively related to nonspecific direction for the hearing impaired, but not for the normal hearing, on 3 of 4 tasks. The direction of the relationship between autonomous achievement motive scores and reward was different for each of the 2 groups for the verbal tasks. The normal-hearing children had higher social comparison achievement motive scores than did the hearing impaired. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Behavior, motivation, and achievement in desegregated junior high school science classes.

To locate possible antecedents for racial differences in science achievement, measures of mathematics and reading achievement, causal attribution, attitude toward school success, and in-class behavior were acquired from 40 Black and 40 White junior high school students, both groups evenly divided between males and females. Significant group differences were limited to mean achievement test scores and science grades. Achievement scores were significantly related to grades, but classroom behaviors also accounted for a significant and unique portion of the variance. There were no significant relationships between behaviors and other student measures. Teacher ratings of effort, ability, course difficulty, and environment were also acquired. Teachers rated Whites as smarter and believed that Black females exerted the greatest effort and Black males the least. As a secondary interest, number and type of student interactions were recorded. Few hostile interracial encounters were observed, and interactions were predominately within race. (18 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Pupil behaviors and interactions in desegregated urban junior high activity-centered science classrooms.

Pupil classroom behavior was observed in 26 Black and 26 White 8–9th graders from 14 activity-centered science classes in 4 urban junior high schools. Classroom behaviors were observed for 10 6-min periods each over 3 mo. 19 behavior categories were coded and later combined to form categories labeled learning, attending, and nonattending. Interpersonal interactions were coded as cross- or within-race and on- or off-task. Entering California Achievement Test mathematics and reading scores (CAT-M and CAT-R, respectively) and final grades were obtained from school records. No significant race differences were found in in-class behavior, few hostile interactions were observed, there were more within- than cross-race interactions, and cross-race interactions were more apt to occur during learning than during nonattending. Reading and math scores were correlated with final grade for both groups. Learning behavior accounted for significant variance in final grade, after removing effects of CAT-M and CAT-R, for Black but not for White Ss. (8 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Culture and test session behavior: Part I.

The relationship between culture (circumscribed by race, ethnicity, country of origin, language, and/or social class) and test session behavior (TSB) on group and individual tests of intelligence and/or achievement is evaluated with respect to the quality of evidence offered in support of knowledge claims. Four relevant categories in the literature are identified: (1) speculative theories, (2) studies of standardized measures of test session behavior, (3) TSB experimental research, and (4) research in cross-cultural test development. In Part I of a two-part series, key literature from the first two categories is reviewed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Incidental learning in adults of different social backgrounds.

Reports unexpected findings in a survey of 191 urban adults from different social backgrounds who were tested on several standard intelligence measures. Social status of all Ss was consistently related to intelligence measures except Picture Order Incidental, a measure of incidental learning. These scores consistently correlated higher with other intelligence measures and with measures of achievement motivation among Ss of low social status than among those of high social status. There was no main effect of race, and none of the intercorrelation patterns of any of the other measures showed this trend. Possible explanations for this anomaly are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Measuring intelligence with the Goodenough-Harris Drawing Test.

Critically evaluates the literature on the Goodenough-Harris Drawing Test (GH) through 1977. Areas reviewed are administration and standardization of the Man and Woman scales, test ceiling, sex differences, the Quality scale, reliability, criterion validity, validity with measures of academic achievement, cultural variables, and use with the learning disabled and the mentally retarded. It was found that although the GH is a reliable measure for children between 5 and 12 yrs old, it is not a valid predictor of criterion measures of intelligence or academic achievement. Socioeconomic status was found to exert a more powerful effect on GH performance than race, geographic location, or size of place of residence. It is recommended that future research focus on the use of the test as a gross screening device for those in the lower ranges of intelligence. (89 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Reciprocal peer tutoring: An analysis of "teacher" and "student" interactions as a function of training and experience.

Investigated the relationship between peer-tutoring interactions of dyads with experience in a reciprocal peer tutoring (RPT) program in mathematics, vs dyads with no experience in RPT. It was hypothesized that students who participated in the RPT condition would exhibit significantly greater amounts of behaviors associated with effective peer teaching than students in the Practice Control (PC) condition and that these behaviors would be associated with mathematics achievement and student reports of behavioral conduct and social competence. 40 academically at-risk African Americans (aged 9.4–12.1 yrs old) from an urban elementary school were randomly assigned to either RPT or PC conditions. Group difference data revealed that RPT Ss displayed significantly higher rates of mathematics achievement, self-report of social acceptance and behavioral conduct and higher rates of observed teacher and student task-related behavior as compared with controls. Peer student active engagement in academic activity was positively related to curriculum-based mathematics test scores. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Academic underachievement, other-direction, and attitudes toward women's roles in bright adolescent females.

Studied achievement patterns and the relationship of other-direction and attitudes toward women to achievement in 43 10th-grade females with IQ scores above 110. Underachievers were Ss whose high school grade average was below the class mean. Achievement indices consisted of high school grades and scores on the Stanford Achievement Test and the Iowa Tests of Educational Development. Grade averages from Grades 2 through 10 were examined. Achievement patterns revealed a significant difference between the grades of achievers and underachievers beginning at Grade 6. Other-direction and attitudes toward women were significantly related to mathematical achievement test scores and high school grades. Results are discussed in relation to research in locus of control and factors contributing to adult underachievement in bright females. (18 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Are SVIB interests correlated with differential academic achievement?

Segel's (1934) finding that SVIB scores correlated more highly with differential academic achievement than with absolute academic achievement has been often quoted, but little studied. The relationship between SVIB scores and ACT test scores for 1875 university freshman males was compared with the relationship between SVIB scores and the differences between pairs of ACT tests. The SVIB scale scores were more highly correlated with differential achievement than with absolute achievement when scholastic aptitude scores were held constant; however, the relationship was slight. When only hypothesized relationships were considered, no difference was found. The interpretation of SVIB scores as reflecting variations in either absolute academic achievement or differential academic achievement should be highly guarded. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The role of IQ and education in predicting later labor market outcomes: Implications for affirmative action.

The authors examine the complex relationships among intelligence, educational attainment, and labor market outcomes. First discussed are differences between psychometric and bioecological approaches to the study of intelligence. Next, the authors describe a study examining the relationship among scores on the Armed Forces Qualification Test (AFQT) taken in adolescence, educational attainment, and the labor market outcomes of wages, personal income, and occupational status 15 years later. At all levels of labor market outcomes, there were greater mean Black–White differences for AFQT scores than for educational attainment. Educational attainment strongly mediated the effects of AFQT scores; there were also substantial racial and gender differences in effects. The article closes by considering the implications of using intelligence test scores and educational attainment as selection criteria. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Are Americans becoming more or less alike? Trends in race, class, and ability differences in intelligence.

American students' test scores have been slowly but steadily declining for the past half century. Some recent explanations for this decline have focused on dysgenic trends resulting from low-IQ parents outbreeding high-IQ parents. In this article, the authors examined the evidence for dysgenic trends by considering race-, class-, and ability-related changes in intelligence test scores over time. They concluded that (a) racial differences in intelligence decreased from 1973 to 1988 and have remained fairly constant since, (b) intelligence differences between the upper and lower thirds of social class groups have been decreasing slightly since 1932, and (c) Preliminary Scholastic Assessment Test-score differences between the top and bottom quartiles have been relatively stable since 1961. Thus, the authors found no evidence supporting the dysgenic hypothesis. Rather, the combined evidence points to a growing convergence across racial, socioeconomic, and ability-related segments of American society. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Examiner observations of children's WISC—R test-related behaviors: Possible socioeconomic status, race, and gender effects.

White examiners completed test observations on 311 children between ages 7 and 14 following the administration of the Wechsler Intelligence Scale for Children—Revised (WISC—R). The children differed by race (White, Black, and Mexican American), social class (middle and lower), and gender. Test observation ratings were employed to predict WISC—R Verbal, Performance, and Full Scale IQs. All 63 correlations between test observations and WISC—R IQs were statistically significant and were reasonably equivalent across gender, social class, and racial groups. Differences in regressions were statistically significant for 23 of the 36 analyses. Although gender differences were not significant, race and social class differences are significant for intercepts, but not for slopes. Thus, given children of the same IQ, White examiners generally observe higher degrees of cooperation, attention, and self-confidence among Black and Mexican American (compared to White) children and among lower-class (compared to middle-class) children. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The role of speech characteristics in scholastic success.

Compared 2 groups of 93 French-Canadian 3rd grade boys in terms of verbal intelligence and various characteristics of speech. Ss in both groups came from low social class homes, and were similar in measured nonverbal intelligence. The essential difference between groups was school achievement; 1 group had average grades of B or above, the other, C or below. It was found that the groups differed reliably in verbal intelligence and in particular speech characteristics, analyzed by an experienced linguist. Distinctive patterns of speech differentiated the 2 groups even with verbal intelligence statistically controlled. Results are discussed in terms of current psycholinguistic theories of social-class differences in speech and language. (French summary) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Personality, problem solving, and the Einstellung effect.

2 hypotheses were raised in this experiment: (a) Ss who show antiscientific or nondynamic tendencies in their perceptions of behavior are more subject to the Einstellung effect; and (b) when achievement anxiety is aroused this relationship is more pronounced. 2 groups of 31 Ss each were given 2 Einstellung-test problem series. For one group standard testing conditions were employed; for the other, achievement anxiety was aroused. Ss were evaluated for dynamic-nondynamic perceptions of others, a measure derived from a modification of Kelly's Role Construct Repertory Test. This measure was not found to be related to Einstellung-test scores, but a significant relationship to achievement anxiety was obtained. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Self-esteem and educational achievement: Independent constructs with a common cause?

Used causal modeling techniques to examine the relationships among social class, ability, educational achievement, and self-esteem for 4 samples of children born in the years 1961–1964. 715 predominately White, urban Ss were involved. Self-esteem was measured through the Self-Esteem Inventory, ability through the WISC, and achievement through the Stanford Achievement Test and Wide Range Achievement Test. Measures were collected when the Ss were between the ages of 4 and 15 yrs. Social class and ability were found to be strongly interrelated and to "cause" both achievement and self-esteem. Achievement was highly stable across the age range 9–25. Results indicate that achievement and self-esteem are not causally related.(32 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Breastfeeding, dummy use, and adult intelligence

BACKGROUND: Several studies show that children who were breastfed as babies gain higher scores on intelligence tests than those who were bottlefed. Although these findings suggest that breastfeeding in early life may promote cognitive development, their interpretation is complicated by the current association between breastfeeding and higher social class. We investigated the relation between method of feeding in infancy and adult intelligence in a setting where breastfeeding was not linked with socioeconomic advantage. METHODS: We followed up 994 men and women, born between 1920 and 1930 in Hertfordshire, UK, for whom information on infant feeding had been recorded by health visitors. Intelligence was measured by the AH4 IQ test, taken on a computer. Factors significantly linked with IQ were investigated by multivariate analysis. FINDINGS: Study participants who had been exclusively breastfed gained slightly higher scores on the IQ test than those who had been exclusively bottlefed, or fed with both breast and bottle. IQ was lower in participants who had used a dummy (pacifier) in infancy, in those whose fathers were in manual occupations at the time of their birth, and in those whose mothers were young at the time they were born. Scores on the IQ test fell as the number of older siblings increased. In multivariate analysis, after adjustment for the effect of all other variables, no association was found between adult intelligence and method of feeding. Dummy use in infancy, number of older siblings, maternal age at birth of the participant, and father's occupational class remained independent predictors of adult intelligence. INTERPRETATION: The mechanisms that link type of feeding in early life with later intelligence may have more to do with the child's social environment that with the nutritional qualities of the milk.     

Analysis of teacher judgments of pupil achievement levels.

Analyzed the accuracy of 12 teachers' judgments of pupil achievement levels. Ss were asked to provide for each of their 322 3rd–8th grade pupils an estimate of achievement test performance, a rating of basic intellectual ability, and a rating of motivation to do school work. Pupils were administered the Gates-MacGinitie Reading Test and the Lorge-Thorndike Intelligence Test. Analyses revealed a high level of accuracy for the achievement judgments when compared to test scores. There was no evidence that pupil gender functioned as a biasing variable in the judgments, but there were indications that the pupil ability variable was a source of bias for some Ss, who overestimated the performance of high-ability pupils and underestimated the performance of low-ability pupils. (20 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Influences on the sociometric ratings of mildly handicapped children: A path analysis.

Hypothesized a model of influence that characterizes teacher and peer perceptions of cognitive and behavioral competencies as mediating variables between observed behavior and achievement and social status. 133 mildly mentally retarded children (aged 7–14 yrs) enrolled in self-contained special classes were observed for attending and disruptive behaviors over 6-sec intervals in round-robin fashion until 10 intervals were collected for each S. Ss were also administered the Wide Range Achievement Test, the Guess Who scale, and a social status measure. Teachers were administered the Teacher Rating scale. Results of a path analysis reveal the following causal sequence or pathway of influence: (a) from observed behavior and achievement, (b) to teacher perceptions of these competencies, (c) to student perceptions of the competencies, and then (d) to social status of the mildly handicapped students. Theoretical and practical implications for the clarification of the dynamics of social status are discussed. (32 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Which is the stronger correlate of school learning? Time to learn or measured intelligence?

Time to learn (TTL; measured by the number of trials to criterion on single units of material from 6 academic areas) and intelligence (measured by the Lorge-Thorndike Intelligence Test) were correlated with each other and with school achievement (measured by standardized test scores) in a study of 153 4th, 5th, and 6th graders. TTL was a strong correlate of school achievement (.85–.89). Correlations were lower between TTL and IQ (.50–.69) and between IQ and achievement (.59–76). It is concluded that TTL is a stronger correlate of school learning than IQ is. TTL is discussed as a tool for teachers who design individual instruction and as potentially the fairest measure of school performance available at the present time. (15 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)