Alphabetic bias in the selection of reviewers for the American Journal of Roentgenology.

OBJECTIVE

This study was performed to determine whether the number of invitations extended to American Journal of Roentgenology (AJR) reviewers is biased toward reviewers with last names that start with early letters of the alphabet.

MATERIALS AND METHODS

The data for this study were extracted from the database of Editorial Manager, the Web-based software used by AJR to manage peer review and other stages of journal production. The alphabetic distribution of last names was extracted from the list of all AJR reviewers in the system between February 1, 2007, and September 13, 2007. The number of reviewer invitations extended to each letter of the alphabet was extracted for the same time period. These data were analyzed using R, a software environment for statistical computing and graphics.

RESULTS

During the 224-day sample period, 1,195 manuscripts were submitted to AJR, and 5,825 invitations were sent to reviewers selected from a pool of 1,573. A linear association was noted between the number of review invitations and the alphabetic position of the first letter of the reviewers' last names, with a downward trend from A to Z (r = -0.75). The observed and expected alphabetic distributions of reviewer invitations were statistically significantly different (chi-square goodness-of-fit test, p < 2.2 x 10(-16)). A reviewer whose last name started with A received a mean of 5.49 invitations during the study period, whereas a reviewer whose last name started with S received a mean of 2.90 invitations. Extrapolating these data to 1 year yields 8.9 and 4.7 invitations to review per year for A and S, respectively--a difference of 4.2 invitations per year.

CONCLUSION

During the study period, there is clear evidence of bias toward reviewers whose last names begin with a letter at the beginning of the alphabet. This bias is both statistically and clinically significant, with reviewers with names starting with letters at the beginning of the alphabet invited to review almost twice as often as those with names starting with letters toward the end of the alphabet. This bias is most likely due to "satisfaction of search" by the assigning editors who tend to invite the first panel of reviewers who meet their criteria on a list of names presented to them in alphabetic order. Reviewers' good will can be enhanced if particular alphabetic groups are not overloaded with reviews. Journal editors can help to avoid this bias by consciously spreading review invitations evenly throughout the whole alphabet. Redesign of editorial workflow software could help to avoid this alphabetic bias.