No possibility of a selection bias, but direct evidence of a simultaneous superiority effect: a reply to Wells et al.

Abstract

The AJS field study was conducted across four different sites (Austin, Texas; Tucson, Arizona; Charlotte-Mecklenburg, North Carolina; and San Diego, California), but nearly 70 % of the lineups were administered at the Austin, Texas site alone. To retain most of the data while eliminating site variance (in an effort to maximize power), our evidentiary strength ratings study was limited to the lineups administered in Austin. We found that suspects identified from those simultaneous and sequential lineups were more likely to be associated with independent evidence of guilt if they had been identified from a simultaneous lineup than from a sequential lineup. In other words, we observed a significant simultaneous superiority effect. Wells et al. (this issue) point out that the simultaneous suspects were also more likely to be adjudicated guilty (77.3 % of simultaneous suspects were adjudicated guilty against 53.3 % of sequential suspects), which they regard as evidence of a sampling bias. However, all suspect ID cases from Austin that could possibly be included in our ratings study were included (i.e., we did not randomly sample a subset of suspect ID cases). Thus, what appears to be a selection bias to Wells et al. is actually further direct evidence of a simultaneous superiority effect.