Thurstone's Primary Mental Abilities Research Articles

A Mixed Method Research Study on the Effectiveness of Board Game Based Cognitive Training Programme

This study aimed to examine the influence of the ‘Board Game Based Cognitive Training Programme’ (BGBCTP) on the cognitive development of the second and third graders among primary school children. BGBCTP is based on educational board games for second and third graders and aims to help them develop cognitive skills. Mixed method was used and it was carried out within the framework of embedded design. A quasi-experimental pretest-posttest design was the quantitative data gathering method and a case study was the qualitative one. The participant group consisted of 120 children (60 in the experimental group, 60 in the control group). The interviews were conducted with twenty grade teachers who carried out the training programme in the experimental group. The General Information Form, The Thurstone Primary Mental Abilities Test 7-11 (PMA 7-11) and “Semi-structured Interview Form” were used to collect data. PMA 7-11 was used to statistically test the influence of BGBCTP on the development of second and third graders’ cognitive skills. BGBCTP was used by the researcher with the children in the experimental group in a ‘regular and controlled’ manner, for 12 weeks, 2 days in a week, 1 hour each day (a total of 24 hours), in addition to their daily activities in their regular environment. For quantitative data analysis, the dependent-samples t-test was carried out for in-group comparisons, whereas the independent-samples t-test was used for intergroup comparisons. For qualitative data analysis, a descriptive analysis approach was used. The quantitative research findings indicated that the BGBCTP has had a significantly positive effect on the cognitive developments of the children in the experimental group. Besides the qualitative research findings revealed that the BGBCTP has a positive effect on children’s linguistic, shape-space, reasoning, discrimination and numerical abilities. In conclusion, BGBCTP is an effective programme on the cognitive development of children.

GENERAL MENTAL ABILITY, NARROWER COGNITIVE ABILITIES, AND JOB PERFORMANCE: THE PERSPECTIVE OF THE NESTED-FACTORS MODEL OF COGNITIVE ABILITIES

The nested-factors model is a well-established structural model of cognitive abilities in cognitive ability research but has not yet been used to investigate the role of cognitive abilities in job performance. Core assumptions of the nested-factors model are that a broad general mental ability (GMA) exists besides narrower abilities and that this GMA differs from the narrower cognitive abilities in breadth but not in subordination. The authors of this article propose that a recently emerging statistical technique—relative importance analysis—corresponds to the assumptions of the nested-factors model. To empirically study the implications of using the nested-factors model, the authors applied relative importance analysis to a meta-analytic matrix linking measures of 7 narrower cognitive abilities from an established ability taxonomy (Thurstone's primary mental abilities), GMA, and job performance. Results revealed that GMA accounted for 10.9% to 28.6% of the total variance explained in job performance and that GMA was not consistently the most important predictor. The discussion focuses on potential theoretical, methodological, and practical implications of the nested-factors model for personnel psychology.

How Impulsivity is Related to Intelligence and Academic Achievement

This study investigated the relationships between impulsivity, intelligence, and academic failure in a sample of 241 secondary school students who completed Thurstone's (1938) Primary Mental Abilities (PMA) test and Dickman's (1990) and Barratt's (1985) impulsivity questionnaires (DII and BIS-10, respectively). Results show an inverse relationship between impulsivity and intelligence, specific to the scales with higher loadings on crystallized intelligence, and a positive relationship between impulsivity and academic failure. These results indicate that impulsivity is not directly related to intelligence and may act as a moderator variable between individuals' resources and their achievements.

Age differences in the structure of intelligence: influences of information processing speed

A battery of widely studied psychometric ability tests, measuring seven primary abilities, was administered to undergraduate students and a cross-sectional sample ranging in age from 43 to 78. The battery included measures of how rapidly individuals could mark answer sheets when provided with booklets containing correct answers to test questions. Confirmatory factor analyses showed that the ability factors could be identified in all age groups, but that the factor structure did not show (metric) invariance of factor loadings across age groups. Factor correlations increased with age, as did the ability tests' communalities, indicating a type of dedifferentiation of the primary ability factor structure. Thurstone's Primary Mental Abilities (PMA) Verbal Meaning Test was shown to have a strong relationship to answer-marking speed, independent of Verbal Ability, and this relationship was higher for older adults. Several ability factors had high correlations with a factor measuring answer sheet speed, and controlling for speed by removing the answer sheet-related variance attenuated the pattern of higher factor correlations for older adults. Findings were consistent with the view that speed of information processing can be both an important correlate of individual differences in rates of intellectual aging and a performance-specific confound that distorts estimates of age-related change in psychometric ability tests.

Perceived Intellectual Performance Change Over Seven Years

Actual and perceived change in intellectual performance over seven years was examined in a sample of 837 participants in the Seattle Longitudinal Study who took five subtests of Thurstone's Primary Mental Abilities (PMA) test in 1977 and 1984. In 1984 participants rated perceived change in intellectual performance from 1977 to 1984. Participants were categorized, based on their actual performance, into those who maintained earlier performance level, significantly increased their performance, or declined in performance. A typology linking actual and perceived change in performance was created: Realists (those who accurately estimated change in their performance); Optimists (those who oveestimated positive change); and Pessimists (those who overestimated negative change). Classification of participants varied across abilities. Women were more likely to be pessimists on Spatial Orientation than men. Older individuals were more likely to be pessimists on Verbal Meaning and Inductive Reasoning Abilities and to be realists on Number ability compared to younger participants.

Relationship between rigidity^flexibility and cognitive abilities in adulthood.

The interrelationship of measures of rigidity-flexibility and of psychometric intelligence is examined. The latent factors of Attitudinal Flexibility, Motor-Cognitive Flexibility, and Psychomotor Speed are derived from the Test of Behavioral Rigidity, and factors of Inductive Reasoning, Spatial Orientation, Verbal Ability, Numeric Ability, Verbal Memory, and Perceptual Speed are derived from the Thurstone Primary Mental Abilities Test and the Educational Testing Service Kit of Factor-Referenced Tests. The data base in this study comes from the fifth wave of the Seattle Longitudinal Study (N = 1,628; age range, 22-95 years). The Rigidity-Flexibility factors were found to be independent of the cognitive domain. Also, longitudinal stability of the factor structure of the rigidity-flexibility domain was confirmed for 837 participants tested in both 1977 and 1984.

Cognitive Benefits of Logo and Computer-Aided Instruction among Zimbabwean Schoolgirls

There is evidence for (2) and against (5) Papert's (4) claim that the educational computing language LOGO fosters cognitive growth. Mundy-Castle and colleagues (3) argue that the social emphasis of LOGO is especially fitting among African children, whose cultures emphasize socially-oriented cognitive skills (1). They found that, in comparison to those who received no computer exposure, black Zimbabwean girls who did LOGO showed significantly greater gains in spatial reasoning. However, the absence of a control group exposed to other computing programmes precluded differentiation between the effects of LOGO and those of computer exposure in general, a limitation this study sought to rectih. One hundred forty-four Zimbabwean primary schoolgirls (M age = 10.4 yr., SD = 1.1) at one school did LOGO for 45 min. a week for one year and 146 girls (M age = 10.4 yr., SD = .9) at a comparable school did computer-aided instructional (CAI) programmes in English, mathematics, and problem-solving for a similar period. Preand posttests were based on Thurstone's (7) Primary Mental Abilities scales (Grades 4-6), which yield verbal, numerical, spatial and perceptual subscale scores. Differences between preand posttest subscale scores for each group were calculated (6) and subjected to multivariate analysis; effects were nonsignificant (F,,,,, = 1.45, p>.05). These data suggest that, notwithstanding potential cultural appropriateness, as taught here, LOGO may be no more effective than CAI in promoting cognitive abilities, at least those measured by standardized inventories.

Assessing intelligence cross-nationally: A case for differential item functioning detection

Cross-national comparisons of intelligence require the use of measurement instruments that are equivalent across cultural groups. One method for insuring equivalent measures is to use only those test items for which the relationship between ability andand probability of a correct response is the same for both national groups. American and German intelligence tests, based on Thurstone's primary mental abilities, were administered to 205 American and 217 German male and female college students in their native languages. Effects of speed were minimized by examining only those items that the majority of subjects reached. Tests were then examined for items that exhibited differetial item functioning ( dif). Group subtest scores were compared before and after dif items and the effects of speed were removed. Ability differences before and after were not the same. The benefits of using a dif analysis and removing the effects of speed in order to establish measurement equivalence in cross-national comparisons of intelligence are discussed.

Egyptian results on the standard progressive matrices

Specific psychometric parameters of the Standard Progressive Matrices (SPM) for Egyptian undergraduate 205 males and 247 females were investigated. The 1 wk retest reliability was 0.82 ( n = 44). A clear and high loaded factor was extracted from the 5 sets, indicating high internal consistency. The factor analysis of the SPM total score and 4 subtests of Thurstone's Primary Mental Abilities, that is, verbal comprehension, number, space, and inductive reasoning, yielded a general and high loaded factor, on which the SPM saturation is 0.77. This denotes significant and high concurrent validity. The mean score of Egyptian males was quite similar to that of British males, however, the mean score for the females was somewhat low. It is concluded that the SPM is a viable tool in the Egyptian context.

Stability and change in adult intelligence: II. Simultaneous analysis of longitudinal means and covariance structures.

We analyzed data on psychometric intelligence from the Seattle Longitudinal Study, simultaneously estimating longitudinal factors, their covariance structure, and their mean levels. Data on five Thurstone Primary Mental Abilities subtests were available for 412 adults, ages 22-70 at first test, who were tested three times at 7-year intervals. A previous longitudinal factor analysis had shown high stability of individual differences (covariance stability) in general intelligence for three adult age groups. We extended that model to estimate factor means. All three age groups showed high levels of covariance stability, but differed sharply in their mean profiles. The young group showed increasing levels of general intelligence, the middle-aged group had stable levels of intelligence, and the old group showed salient, approximately linear, decline. The patterns of stability in middle-age, followed by mean decline and high covariance stability in old age, suggest a normative developmental transition from a stability pattern to a decline pattern of general intelligence, with the inflection point occurring somewhere around age 60.

Stability and change in adult intelligence: 1. Analysis of longitudinal covariance structures.

We address two questions of central interest in adult intellectual development: the equivalence of psychometric tests' measurement properties at different ages, and the stability of individual differences in intelligence over time. We performed a series of longitudinal factor analyses using the LISREL program to model longitudinal data from Schaie's Seattle Longitudinal Study. The results indicate complete invariance in the loadings of five subtests of Thurstone's Primary Mental Abilities battery on a general intelligence factor. Individual differences in general intelligence were highly stable over 14-year epochs, with standardized factor correlations averaging about .9 between adjacent 7-year testing intervals. These results indicate that most individuals in this relatively select longitudinal sample maintained their relative ordering in intelligence.

Hierarchical Structure of Abilities: Factorial Validation of Higher Order Constructs in Thurstone's Primary Mental Abilities

Vernon and Humphreys have advocated the use of the hierarchical model as a means of conceptualizing the organization of human abilities and of thinking about tests. In this study, Thurstone's primary mental abilities (1938/1968) involving some 57 tests were factor analyzed to produce a comprehensive hierarchical model. Thurstone's study of primary mental abilities was designed for the measurement of individual differences and the development (or the use) of tests. Applying the hierarchical model, a test constructor might be able to select, according to his needs for measurement, factors or constructs at any level in the hierarchy.

A CHECK ON THE THEORY OF FLUID AND CRYSTALLIZED INTELLIGENCE WITH DESCRIPTION OF NEW SUBTEST DESIGNS

Several important theories about human abilities contend in the field today. Salient features of these theories have been explicated and critically evaluated in several recent publications. For example, Horn (1972, 1976, 1977a, 1977b, 1978a, 1978b) has considered the evidence in relation to the classical theories of general intelligence, Thurstone's primary mental abilities structure, the hierarchical theories of the British psychologists (Burt & Vernon) and Humphreys, the Piagetian system, Guilford's structure of intellect model, Guttman's analytic/achievement distinction, and Jensen's level I/level II conception. Matarazzo (1972) has looked at much of this evidence from the position of one presenting a new handbook for the Wechsler scales. Cattell (1963, 1971) and Horn (1968, 1970, 1972, 1974, 1975, 1977a, 1978a, 1978b) have evaluated the evidence in terms of theories of fluid (Gf) and crystallized (Gc) intelligence. Several recent efforts have been directed at linking together the evidence and theory deriving from study of information processing or cognitive psychology, on the one hand, and the study of individual differences, on the other hand (Carroll, 1976; Horn, 1978c; Hunt, 1978; Sternberg, 1977). The hypotheses of these theories that pertain to organization among abilities are of major importance for understanding the present study. The classical viewpoint (Spearman-Binet-Wechsler) is still quite influential, particularly in applied settings. The major concept of this kind of theory is a single general factor of intelligence comprised of a collection of abilities. A concept of general intelligence is also an important feature of Piagetian theory, but in place of a collection of abilities this theory postulates a collection of schemata, these being cognitive laws in accordance with which a person explains or understands his conception of reality. In Guilford's model the domain of behavior represented by a general intellective factor is described in terms of a large number of distinct abilities said to be indicated by uncorrelated factors. This theory can be viewed as one of several efforts to deal with the proliferation of ability factors presented to the literature since the modern computer came on the scene (Humphreys, 1962). Other notable

A method for analyzing the genetic basis of covariation.

Covariances between traits can be partitioned into additive and dominance genetic components and between- and within-family environmental components, using a method analogous to that used in the analysis of single traits. The problem arises as to whether all additive genetic components simply reflect a single additive component, in the sense that, given an appropriate rescaling of the breeding values, a single additive genetic component would adequately describe the additive genetic variation. The statistical procedure for testing this hypothesis is discussed in detail. Similar considerations apply to the dominance variation. The approach is applied to twin data given by Loehlin and Vandenberg (1968) on covariation between five of Thurstone's Primary Mental abilities. Although the data do not permit a reliable separation of additive and dominance components, it is shown that a single genetic component will account for almost all the genetic variation and covariation. Unless there is marked linkage disequilibrium, this implies that most of the genetic variation for the five traits can be attributed to the pleiotropic action of genes at a common set of loci.

On the Internal and External Validity of Simple Longitudinal Designs

LABOUVIE, ERICH W.; BARTSCH, THOMAS W.; NESSELROADE, JOHN R.; and BALTES, PAUL B. On the Internal and External Validity of Simple Longitudinal Designs. CHLD DEVELOPMENT, 1974, 45, 282-290. The effects of selective dropout and retesting on the assessment of ontogenetic changes in abilities during adolescence are examined. A stratified random sample of 1,580 adolescents (cohorts 1953-1957) was measured in 1970 with Thurstone's Primary Mental Abilities Test, yielding scores on 6 ability dimensions. According to participation 1 year later, in 1971, the original sample was subdivided into 1,214 retestees and 366 retest resisters. In addition, a random stratified sample of 263 control subjects was measured only once in 1971. Results indicate that volunteering behavior of adolescents, being correlated with various indices of intellectual functioning, jeopardizes the external validity of longitudinal designs. Moreover, different determinants of nonvolunteering (refusers vs. nonavailablle participants) may result in differential biases of the remaining longitudinal sample. In addition, the findings indicate that age-related longitudinal increases on intelligence variables are mainly due to retest effects.

Thurstone's primary mental abilities and structure-of-intellect abilities.

Thurstone's primary mental abilities and structure-of-intellect abilities.

A CLARIFICATION OF CROPLEY AND MASLANY'S ANALYSIS OF THE WALLACH‐KOGAN CREATIVITY TESTS

In a recent article Cropley & Maslany (1969) examined the relationships among and between the Wallach–Kogan creativity tasks and the Thurstone Primary Mental Abilities tests in a sample of university students. A factor analysis yielded a general factor with moderate to substantial loadings for all of the tests in the battery. On this basis Cropley & Maslany concluded that complete statistical independence of the ‘creativity’ and ‘intelligence’ domains had not been established. The present paper attributes the Cropley–Maslany outcomes to a failure to rotate, and reports a Promax rotation of the Cropley–Maslany principal‐components solution. Such rotation yields a pure ‘creativity’ factor, and two ‘intelligence’ factors. Unlike the Cropley–Maslany principal‐axis factor loadings, the Promax solution is congruent with most of the evidence reported in other relevant investigations of the creativity–intelligence distinction.

Innate Abilities, One or Many? A New Method and Some Results

SUMMARY AND CONCLUDING REMARKSA generalization of the F test is proposed to see how many independent components are statistically significant when the co-variances between DZ twin differences on several psychological tests are compared with similar co-variances for MZ twins. When applied to the 6 scores of Thurstone's Primary Mental Abilities test battery, at least 4 separate components were found to be significant independently. Then 4 components are tentatively identified as similar to, although not identical with, the Number, Verbal, Space and Word fluency scores. While the procedure bears some resemblance to factor analysis, it may lead to quite different results since it is based on the comparison of two co-variance matrices of twin differences.In conclusion, it should be kept in mind that these results are based on small samples of MZ and DZ twins, so that a repetition of the study seems indicated. This replication is now in progress.

Varimax solution for primary mental abilities

The varimax solution for Thurstone's classic Primary Mental Abilities study is presented. Comparisons between the factors of Thurstone's original subjectively rotated factor pattern, Zimmerman's subjectively revised solution, Wrigley, Saunders, and Neuhaus' quartimax results, and the present varimax factor matrix are made by finding correlations between factors defined by these four solutions. It is pointed out that any possible ultimate merit of the varimax solution should be based on its psychological meaningfulness and on the rationale of the varimax criterion—not on its relationship to the other studies.

Application of the quartimax method of rotation to Thurstone's primary mental abilities study

This study compares a quartimax rotation of the centroid factor loadings for Thurstone's Primary Mental Abilities Test Battery with factorings of the same correlation matrix by Thurstone (simple structure), Zimmerman (revised simple structure), Holzinger and Harman (bi-factor analysis), and Eysenck (group factor analysis). The quartimax results agree very closely with the solutions of Holzinger and Harman and of Eysenck, and reasonably well with the two simple structure analyses. The principal difference is the general factor provided by the quartimax solution. Reproduction of the factorial structure is sufficiently good to justify its use at least as the first stage of rotation. More extensive trial of the method will be needed with more varied data before it will be possible to decide whether quartimax factors meet psychological requirements sufficiently well without further rotation.