SURVEY INSTRUMENT

Respondents viewed the constituent e-mail experiment first, then the list experiment.

Constituent E-mail Experiment

All respondents viewed the introductory text first, then were randomly presented with three messages: one each about policy, service, and allocation. The constituent's name was also randomized to signal gender and race (white, black, Latino/a). Recall that the data presented in the main text include only the first treatment message each respondent viewed. I report results from this between subjects design rather than the within-subjects design to avoid problems from respondents learning the aim of the study as they viewed and responded to the three messages in succession.

Introductory Text

You will now see a series of three e-mail messages sent to you from hypothetical constituents. In each one, first read the message. Then answer two questions about each one. First, evaluate the priority level you or your office would give this e-mail if it were sent to you today. Select your evaluation by clicking on a number on the “priority scale” under the message. This measure ranges from 0 to 100, with 0 indicating lowest priority and 100 indicating highest priority. If you would give the message average priority, select the score 50. Second, evaluate how likely you would be able to satisfy the constituent.

Policy Treatment

1. • Male names: Connor Fredericks, Tyrone Baker, Juan Lopez

2. • Female names: Molly Wollsteiner, Shanice Jackson, Camila Garcia

Service Treatment

1. • Male names: Jacob Nichols, DeAndre Moore, Antonio Ruiz

2. • Female names: Amy Schneider, Tasheka Robinson, Gabriela Mendez