Desempenho de Alunos do Quarto Ano em Testes de Subitização e Estimativa e no Teste de Desempenho Escolar (TDE)

• Autor(a) principal: Do Prado, Paulo Sérgio Teixeira [UNESP]
• Data de Publicação: 2015
• Outros Autores: De Freitas Betetto, Mariana, Casali-Robalinho, Ivana Gisel, Fioraneli, Rogério Crevelenti, Mendes, Alessandra Campanini, Rocca, Júlia Zanetti, Da Silva Cacciacarro Lincoln, Leila Estevão, Zampieri, Marília, Dos Santos Carmo, João
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.9788/TP2015.1-01; http://hdl.handle.net/11449/227983
• topic: Mathematical psychology; Number sense; Numerical aptitude; School Performance; Statistical correlation

The paper reports a research conducted with 62 students enrolled in the fourth grade of two schools of different cities of São Paulo state: one private and another public. Its purpose was to investigate the relation between their performance on probes of subitizing and estimation and on the School Performance Test (TDE). Subitizing and approximate numerical estimation are part of number sense, which is considered innate, therefore not subject to infl uence of environmental factors such as type of school attended, contrary to what would be expected with respect to math skills learned in school. Additionally, once number sense is innate it would be restricted to limits whose overcoming is made possible by cultural advancements. So between it and math skills one would expect a weak or absent correlation. Results of subitizing and estimating test showed the same pattern of performance found in the literature: virtually no errors with numerosities 1 to 4 and an abrupt drop from 5 to 10. Comparisons were performed according to administrative dependence of schools, age and sex of the participants. No statistically signifi cant differences were found. The results of the arithmetic subtest of the TDE indicate a better performance of private school students, with no differences in respect to the other variables. Finally, the analysis of Correlation (by Pearson) between the data from the two tests showed a moderate to low coeffi cient. Therefore, the results confi rm the hypotheses tested and provide additional evidence in favor of the thesis of an innate numerical sense.