A Problem with Double-Blind Photospread Procedures: Photospread Administrators Use One Eyewitness's Confidence to Influence the Identification of Another Eyewitness

In Experiment 1, photospread administrators (PAs, N = 50) showed a target-absent photospread to a confederate eyewitness (CW), who was randomly assigned to identify one photo with either high or low confidence. PAs subsequently administered the same target-absent photospread to participant eyewitnesses (PWs, N = 50), all of whom had viewed a live staged crime 1 week earlier. CWs were rated by the PAs as significantly more confident in the high-confidence condition versus low-confidence condition. More importantly, the confidence of the CW affected the identification decision of the PW. In the low-confidence condition, the photo identified by the CW was identified by the PW significantly more than the other photos; there was no significant difference in photo choice in the high-confidence condition. In spite of the obvious influence exerted in the low-confidence condition, observers were not able to detect bias in the photospread procedures. A second experiment was conducted to test a post-hoc explanation for the results of Experiment 1: PAs exerted influence in the low-confidence condition because they perceived the task as more difficult for the eyewitness than in the high-confidence condition. Independent observers (N = 84) rated the difficulty of the confederate's task as higher in the low-confidence condition compared with the high-confidence condition, suggesting that expectations of task difficulty might be driving the effect observed in Experiment 1. Results support recommendations for double-blind photospreads and emphasize that the same investigator should not administer photo lineups to multiple eyewitnesses in an investigation.     

Is It Race,Class, or Gender? The Sources of Perceived Discrimination in Brazil

Observers have long noted Brazil's distinctive racial politics: the coexistence of relatively integrated race relations and a national ideology of “racial democracy” with deep social inequalities along color lines. Those defending a vision of a nonracist Brazil attribute such inequalities to mechanisms perpetuating class distinctions. This article examines how members of disadvantaged groups perceive their disadvantage and what determines self‐reports of discriminatory experiences, using 2010 AmericasBarometer data. About a third of respondents reported experiencing discrimination. Consistent with Brazilian national myths, respondents were much more likely to report discrimination due to their class than to their race. Nonetheless, the respondent's skin color, as coded by the interviewer, was a strong determinant of reporting class as well as race and gender discrimination. Race is more strongly associated with perceived “class” discrimination than is household wealth, education, or region of residence; female gender intensifies the association between color and discrimination.     

Where Are the Women? An Analysis of Gender Mainstreaming in Introductory Political Science Textbooks

Textbook content is a powerful indicator of what is and is not considered important in a given discipline. Textbooks shape both curriculum and students’ thinking about a subject. The extant literature indicates that gender is not well represented in American government textbooks, thus signaling to students that women and gender are not part of the mainstream in political science. I contribute to this literature by using quantitative and qualitative content analysis to examine gender mainstreaming in 10 introductory political science textbooks. I find that the quantity of gendered content is small, and the quality of that content varies considerably from text to text.