Analysis of multivariate profiles for the detection of students with ADHD

Abstract

This article presents the results of the analysis of k-means conglomerates to identify multivariate assessment profiles for identifying children with ADHD. To do this, we used the Magellan Scale (EMA-DDA), which requires the evaluation of students by two groups: parents and teachers. The studies that have been carried out in this regard by other authors, although differing in the scales used, show that the nomination made by both groups usually disagree and are not usually concurrent in several of the dimensions that are evaluated. In a previous study we found that there were differences in several dimensions, from the point of view univariate. However, considering that the identification is based on thresholds that allow the classification of students, we wonder if these differences would be observable also multivariate level between the two groups. The interpretative hypothesis we assume, like others before, is based substantially on the two collective value to students in different contexts. This perspective could support, to exploratory level, if that assessment can be the basis of differences, and thus, to guide subsequent esperimental or quasi-experimental validation studies. The study allows us to affirm that there is a relationship between the multivariate profiles that result from the assessments made by the two groups, which supports the interpretation line noted above.