Table 2 Comparison of outlying participants and errors identified by the multivariate outlier detection approaches during the first iteration of the data evaluation process: first, between MCD and RPCA directly, combining results with and without adjustment; then, between the adjusted and unadjusted results within each multivariate method. For each set of results, the total number of outliers/errors by each approach is reported (MCD vs. RPCA; adjusted vs. unadjusted), as well as the number that overlapped between the two approaches

From: The utility of multivariate outlier detection techniques for data quality evaluation in large studies: an application within the ONDRI project