RELATIONS BETWEEN ATTENTIONAL STRUCTURE AND ATTENTIONAL FUNCTION: UTILIZATION OF ALTERNATIVE STATISTICAL APPROACHES

Limited research findings regarding structure-function relations in the domain of attention may stem from problems in estimating these relations in small samples combined with data distributions that do not conform to the assumptions of the statistics used to estimate the relations. We examined the utility of using alternative statistics to estimate those relations. One hundred eleven children (82 spina bifida, 29 normal controls) were included to estimate behavior-behavior relations, and 61 children (43 spina bifida, 18 normal controls) were included to estimate structure-function relations. We used the Pearson’s Correlation and four robust correlations: the Percentage Bend Correlation, the Winsorized Correlation, the Skipped Correlation using the Donoho-Gasko Median, and the Skipped Correlation using Minimum Volume Ellipsoid Estimator to investigate behavior-behavior and structure-function relations in the domain of attention. A bootstrap sampling process was used to compare performance of the five estimators in this field context. The results of the study suggest that utilization of robust methods to estimate structure-function and behavior-behavior relations can assist investigators when confronted with small samples and multivariate non-normal data. Furthermore, the similarity of estimates across correlational methods suggests that the lack of structure-function relations found in the literature is not easily attributed to violations of distributional assumptions.