Correction to: Adv in Health Sci Educ (2017) 22:327–336 DOI 10.1007/s10459-016-9739-9

In re-examining the paper “CASPer, an online pre-interview screen for personal/professional characteristics: prediction of national licensure scores” published in AHSE (22(2), 327–336), we recognized two errors of interpretation.

The first relates to the comparison of CASPer and MMI. In the paper, we set up a contrast between CASPer and MMI, implying the relative worth of each. On page 333, we discussed correlations between CASPer Video Scenarios, MMI, and the MCCQE Part 1 and 2. We contrasted the correlation between CASPer and CLEO (0.36, 0.50) and MMI and CLEO (0.37, 0.44) for Part 1 and Part 2, respectively. As the CASPer-CLEO comparison used a disattenuated correlation whereas the MMI-CLEO comparison did not, the two correlations are not commensurate. Some statements in the paper imply a competition between the two testing strategies. Instead, our current thinking is that the two methods are complementary, with the appropriate role of CASPer as a screening test to bring applicants to an MMI. Moreover, the strategy of examining the relation between components of the CASPer test and other measures may be a promising direction for future research.

The second relates to the correlation between Patient Interaction (PI) score on the Part 2 MCCQE and MMI (0.65; Eva et al. 2009) versus CASPer (0.17). In the paper, we noted a “positive trend” in describing the CASPer–PI correlation. The correlation was not significant and should not be labelled as a “trend”. However, this lack of correlation is informative, in that the PI score is a measure of oral communication, which may be more directly assessed by the MMI than CASPer. Further research might more systematically explore the underlying dimensions or abilities assessed by the two methods.