Comparison of methods in the identification of secondary school students with high abilities

Enfoque

In the field of giftedness, the procedures for identifying gifted and talented students are probably the most discussed and widely written-about topics. The present study primarily aimed to identify high-ability students in a sample of Spanish middle school students. Specifically, the participants were a total of 1456 students of first and second year of Compulsory Secondary Education in the province of Alicante (Spain); however, some students were not included in the final sample because of insufficient language proficiency, incomplete tests, or lack of family consent. Thus, the final sample included a total of 1400 participants (n= 1400). Of this final sample, 52% were first-year students, and the remaining 48% were in their second year. In terms of gender, 52.8% were boys and 47.2% girls. The age range of the participant sample was between 12 and 15 years (M= 12.8, SD = 0.67). In addition, 81.4% of the students belong to a public school and 18.6% to a private one. The measures used were a non-verbal general intelligence test (Factor “g”), a differential aptitude test (BADYG), a creativity test (fluency, flexibility, originality, elaboration, title, and special details), and the average performance measure. The average time for the test administration was three hours and forty minutes distributed over three days. Data analysis techniques varied according to the established objectives and included the following: a) identification method for students with high abilities, for which different cut-off points were used for each of the variables (BADYG, G Factor, Creativity, and Performance)  separately (from the top 10%) and the combined variables, taking into consideration all the variables used in the identification (from broad percentiles such as 90%) (the “or” rule), and an identification method in which an average of the criteria scores exceeding a cut-off point for the top 10% was used (the “mean” rule); b) techniques for verifying agreement among the results of different selection criteria, such as chi-square (c²), phi (j, kappa (k), and intraclass correlation coefficients for continuous variables; and c) means for comparisons for two (student t test) or more (one-way ANOVA) groups. The gifted students were identified using two methods: a) the “or” rule: obtaining a score corresponding to the 90th percentile or higher on any of the tests independently, and b) the “mean “rule: presenting an average of the criteria scores exceeding a cut-off point of the top 10%. The results indicated that the prevalence of students with high abilities is similar to the prevalence identified in other studies when using unique identification criteria. However, the prevalence rate reduced considerably when two or more criteria were combined. The results indicate the need to use different identification criteria and to analyze the heterogeneity existing within the group of high-ability students. Such results would aid in designing educational interventions suitable to the profiles of the different subgroups. The correct identification of gifted students will equip such students to obtain the help, resources, and optimum educational experiences necessary for their optimal development.