Time-sharing Experiments for the Social Sciences, TESS2047 Thornton, Understanding the Role of Religious Appeals in Political Communication

Data Archive > Other Data > Summary

TESS conducts general population experiments on behalf of investigators throughout the social sciences. General population experiments allow investigators to assign representative subject populations to experimental conditions of their choosing. Faculty and graduate students from the social sciences and related fields (such as law and public health) propose experiments. A comprehensive, on-line submission and peer review process screens proposals for the importance of their contribution to science and society.

The following 2X2 experiment examines the concept of religious priming as well as the potential for political awareness to moderate discernible priming effects. The study follows a 2 (Religious Cues: Present, Absent) x 2 (Prior Information: Present, Absent) between subjects factorial design, with pretest and posttest questions. Religious cues are manipulated by providing respondents with a political advertisement including or excluding religious appeals. The second factor manipulates awareness, specifically how much information participants know about the political candidate’s policy preferences. As such, participants will be randomly assigned to one of four experimental conditions (Information-Present, Cue-Present; Information-Present, Cue-Absent; Information-Absent, Cue-Present; Information-Absent, Cue-Absent). The information condition is simply whether one receives a one page pdf bio of the candidate; the religious cue condition is whether one receives a political ad with or without religious cues.

Data File
Cases: 562
Variables: 54
Weight Variable: WEIGHT
Data Collection
Date Collected: 5/18/2010-9/8/2010
Original Survey (Instrument)
TESS2 047_Thornton_FINAL
Funded By
National Science Foundation
Collection Procedures
All participants will complete the 11-item pretest questionnaire (see original survey instrument). Following the pretest but prior to message exposure, half of the participants will be provided with information about the ad-targeted political candidate, which is meant to approximate one’s degree of awareness and sophistication (Manipulation #1). The information will be presented to participants as a one page, modified website taken from the candidate’s 2008 website. Subjects will then view a political advertisement of the candidate either including or excluding religious appeals (Manipulation #2). Specifically, one half of the subjects will be exposed to the complete commercial in its original form (59 seconds), while remaining participants will be exposed to the commercial with religious cues removed (52 seconds). Save for these modifications, the ads are equivalent. All subjects will then complete the nine-item posttest survey (see original survey instrument).
Sampling Procedures
TESS provides investigators an opportunity to run Internet-based experiments on a random, probability-based sample of the population. To achieve a representative sample, we contract with GfK (formerly Knowledge Networks), which conducts surveys using its KnowledgePanel. KnowledgePanel is a nationally representative, probability-based web panel based on dual-frame sampling that combines traditional random-digit-dialing telephone surveying techniques with an address-based technique that allows the sample to be representative of cell-phone-only households as well as those with land-lines. A summary of the KnowledgePanel survey design used for the TESS projects can be accessed here. Additional data and study materials can be downloaded here.
Principal Investigators
Christopher Weber, University of Arizona
Matthew Thornton, Louisiana State University
For any additional information regarding TESS data, please visit the original website here

Bookmark and Share