The present dissertation is concerned with language aptitude in young learners. The first aim of this dissertation is to validate the content of two new instruments: the adaptation to both Spanish (MLAT-ES) and Catalan (MLAT-EC) of the Modern Language Aptitude Test – Elementary (MLAT-E) (Carroll & Sapon, 1967). The second aim of this dissertation is to study the relationship between language aptitude, L1 acquisition and cognitive development. The norming study of the MLAT-E offers data of children from grade 3 to 6 while that of the MLAT-ES covers grades 3 to 7. In both norming studies, while means increase steadily from grades 4 to 6, the largest increase is observed between grade 3 and 4. The third aim of this dissertation is related to the variable sex. While in the MLAT-E study the data appear divided according to this variable, the population in the MLAT-ES norming study appears as one cohort. It would be interesting to see if significant differences are observed between the performance of boys and girls to keep this variable or not when using these tests. Finally, the fourth aim of this study is to check the construct validity of the tests. The MLAT-E tests are intended to measure language aptitude for foreign language learning. One way to check their construct validity is to correlate scores on the tests with measures of foreign language proficiency. The first aim of the study - the validation of two versions of the MLAT-E in Spanish and Catalan - is supported by the statistical results obtained although some of the items in the MLAT-ES do not work as expected (or as they do for monolingual Spanish speakers or speakers of a Spanish regional variety other than the Peninsular). As for the second aim, it was observed that the tests functioned in a different way for 3-graders than for the rest of grades, as in the previous norming studies. This could be explained by the fact that 3-graders are at the beginning of their concrete operational stage, which limits their cognitive processing abilities. From an information processing perspective, children at this stage still have to develop problem-solving strategies as well as strategies for encoding and memorising information. Moreover, they have become literate relatively recently and show to have lower meta-linguistic awareness than their older counterparts. The third aim dealt with the sex variable. The results show that aptitude does not seem to be affected by the participants’ sex. However, perhaps a more cognitive oriented approach into the way each of the parts of the MLAT-E is answered would shed light on sex differences, if any. These should be explored along with different proficiency measures, those that, traditionally, have given an advantage to girls over boys (such as fluency) and vice versa (e.g. receptive tasks). Regarding the construct validity of the tests, both the MLAT-ES and the MLATEC seem to be valid measures of predictability of general concurrent proficiency although not in all skills, since the speaking skill sometimes did not correlate with the aptitude measures. Also, correlations in grade 3 were consistently non-existent or lower than those in other grades. This phenomenon could be related to the issues discussed in relation to the participants’ age and cognitive development and/or to the validity of the proficiency measures used. Consequently, the MLAT-ES and the MLATEC should be revised if they are to be administered to children this age. Oral data were collected from most participants, so perhaps a deep analysis of these oral data will shed more light on the relationship between FL aptitude in young learners and their true FL proficiency in further research. / Aquesta tesi doctoral presenta la validació de dos tests d’aptitud per a l’aprenentatge de llengües estrangeres en un context bilingüe català/castellà. Es tracta de l’adaptació al castellan (MLAT-ES) i al català (MLAT-EC) del Modern Language Aptitude Test – Elementary (MLAT-E) (Carroll i Sapon, 1967). 629 estudiants d’edats compreses entre els 8 i els 13 anys (de 3r de primària a 1r d’ESO) participaren en aquest estudi. Un cop validat el contingut d’ambdós tests, la primera pregunta de recerca explora la relació entre l’aptitud, l’adquisició de la L1 i el desenvolupament cognitiu, ja que els resultats mostren que, si bé les mitjanes obtingudes en els tests augmenten amb l’edat, aquest augment és molt més gran entre 3r i 4t de primària que entre els altres cursos. Aquesta diferència es podria explicar per l’estat de desenvolupament cognitiu dels participants a 3r de primària així com pel fet que en aquesta edat algunes estratègies de resolució de problemes i de memorització, entre altres, encara estan en vies de desenvolupament. La segona pregunta d’investigació té a veure amb la variable sexe. No s’observen diferències significatives entre els resultats obtinguts pels nens i nenes participants. Finalment, l’objectiu de la tercera pregunta de recerca és comprovar la validesa de constructe del MLAT-ES i del MLAT-EC correlacionant els resultats obtinguts amb proves de proficiència en llengua estrangera (anglès). Ambdós tests semblen ser vàlids en aquest aspecte tot i que amb excepcions significatives al grup de 3r de primària, sobretot pel que fa a la proficiència oral. Aquestes diferències es podrien explicar per les proves de proficiència utilitzades o per l’estat de desenvolupament cognitiu d’aquests participants, que podria haver influït en la manera com interpretaven els tests d’aptitud. Per tant, en futurs estudis s’hauria d’explorar amb més detall si les habilitats mesurades pels tests d’aptitud en nens i nenes d’entre 8 i 9 anys són les mateixes que en estudiants més grans.

The present study sets out to evaluate the effectiveness of a dynamic approach to aptitude testing. It was proposed that it is not always appropriate to use conventional aptitude tests to predict future academic success in the South African context. The study posited the belief that an alternative testing format could be facilitated by using a test-train-test procedure within a learning potential paradigm. The learning potential paradigm as formulated through Vygotskian and Feuersteinian theory is operationalised in the form of a Newtest Battery. The Newtest procedure is in direct contrast to traditional approaches to aptitude testing. The latter approaches both implicitly and explicitly adopt a static view of ability, whereas the Newtest approach focuses on the learning potential of the testee, as well as consequent performance. However, the assessment of learning potential poses problems of its own. Modifications were introduced to ensure that the Newtest format is both appropriate and psychometrically defensible. The construction and evaluation of the Newtest Battery is described. A sample of both advantaged and disadvantaged students were tested on a battery of traditional aptitude tests. This group of students was contrasted with another sample of both advantaged and disadvantaged students who undertook the Newtest Battery in the modified dynamic testing format. The traditional measures of aptitude were found to be invalid predictors of university success. Matric results showed a relationship with academic success for both groups. The Newtest measures enhanced the prediction of academic success for both advantaged and disadvantaged students. The Deductive Reasoning dynamic measure was found to be a valid predictor of university success for the disadvantaged students. The results thus successfully extend the learning potential paradigm into the realm of group aptitude testing. The validity of traditional aptitude test measures has been brought into question by the findings of the study. The study points the way forward to a more equitable and relevant aptitude testing procedure. Finally, it was shown that the testing environment forms part of the socio-educational context. Personnel involved in the administration of aptitude tests are given guidelines \vi th the aim of equalising the test process. / Psychology of Education / D. Ed. (Psychology of Education)

The present study sets out to evaluate the effectiveness of a dynamic approach to aptitude testing. It was proposed that it is not always appropriate to use conventional aptitude tests to predict future academic success in the South African context. The study posited the belief that an alternative testing format could be facilitated by using a test-train-test procedure within a learning potential paradigm. The learning potential paradigm as formulated through Vygotskian and Feuersteinian theory is operationalised in the form of a Newtest Battery. The Newtest procedure is in direct contrast to traditional approaches to aptitude testing. The latter approaches both implicitly and explicitly adopt a static view of ability, whereas the Newtest approach focuses on the learning potential of the testee, as well as consequent performance. However, the assessment of learning potential poses problems of its own. Modifications were introduced to ensure that the Newtest format is both appropriate and psychometrically defensible. The construction and evaluation of the Newtest Battery is described. A sample of both advantaged and disadvantaged students were tested on a battery of traditional aptitude tests. This group of students was contrasted with another sample of both advantaged and disadvantaged students who undertook the Newtest Battery in the modified dynamic testing format. The traditional measures of aptitude were found to be invalid predictors of university success. Matric results showed a relationship with academic success for both groups. The Newtest measures enhanced the prediction of academic success for both advantaged and disadvantaged students. The Deductive Reasoning dynamic measure was found to be a valid predictor of university success for the disadvantaged students. The results thus successfully extend the learning potential paradigm into the realm of group aptitude testing. The validity of traditional aptitude test measures has been brought into question by the findings of the study. The study points the way forward to a more equitable and relevant aptitude testing procedure. Finally, it was shown that the testing environment forms part of the socio-educational context. Personnel involved in the administration of aptitude tests are given guidelines \vi th the aim of equalising the test process. / Psychology of Education / D. Ed. (Psychology of Education)