STEREOTYPE PRIMING THROUGH NEWS STORY HEADLINES: USE OF THE WORD TERRORIST TO PROMPT IMPLICIT ASSOCIATIONS WITH MUSLIMS
 
Student researcher

Jennifer Hoewe (graduate student)
Brett Sherrick (graduate student)
Alyssa Appelman (graduate student)

This paper was based on a project as part of the Comm 506 course

Faculty Supervisor

Dr. S. Shyam Sundar

Introduction

This study sought to determine if a stereotype prime in a news story headline could activate implicit stereotyping, specifically regarding the Muslim terrorist stereotype. Also, this study developed and utilized a new measure for research on implicit stereotype priming, called a hypothetical statement decision task, which was recommended for future use.

RESEARCH QUESTION(S) AND HYPOTHESES:

This study examined the role of news media consumption as a moderating variable. A high level of exposure to news media should mean individuals have encountered more instances of the Muslim terrorist stereotype. Therefore:

H1: If primed with the word terrorist in the headline of a news story, participants with greater amounts of news media consumption will be more likely to associate Muslims with negative hypothetical statements than participants with lesser amounts of news media consumption.

H2: If primed with the word terrorist in the headline of a news story, participants with greater amounts of news media consumption will be less likely to associate Muslims with positive hypothetical statements than participants with lesser amounts of news media consumption.

Religious affiliation and level of religiosity also were examined to determine what, if any, moderating influence these variables have on the accessibility and application of the Muslim terrorist stereotype. Therefore:

RQ1: What influence do religious affiliation and level of religiosity have on participants’ willingness to associate Muslims with negative or positive hypothetical situations, if they are primed with the word terrorist in the headline of a news story?

Method

Sixty undergraduate students were recruited from communications courses to participate in a between-subjects experiment. To accommodate the stereotype prime, the headline of the news story was determined by the participant’s condition. In the experimental condition, the headline contained the word terrorist; whereas, the headline in the control condition did not. The body of the news story was identical for both conditions. Participants were presented with a news story and then instructed to complete the hypothetical statement decision task, consisting of 20 positive and negative hypothetical statements appearing in a different random order for each participant.

Results

A paired t-test was conducted to test whether, across conditions, an anti-Muslim stereotype emerged. It showed that participants associated more negative statements with Muslims (m = 4.25, SD = 0.83) than they associated positive statements with Muslims (m = 0.77, SD = 0.83). This difference was statistically significant at two-tailed t(59) = 8.53, p < .001.

A general linear model examining negative associations with Muslims did not identify a statistically significant difference between conditions, F(1,56) = 0.47, p = 0.49. Another general linear model on negative associations with Muslims showed that news consumption had no main effect, F(1, 56) = 1.52, p = .22. The interaction between condition and news consumption for negative hypothetical statements also was not significant, F(1, 56) = 0.07, p = .79. Therefore, H1 was not supported.

A general linear model examining positive associations with Muslims did not identify a statistically significant difference between conditions (when amount of news consumption was not addressed), F(1,56) = 0.92, p = 0.34. Another general linear model on positive associations with Muslims showed that news consumption had no main effect when considering all participants, F(1, 56) = 2.34, p = .13. However, for participants in the control condition, there was a positive correlation between amount of news consumption and number of positive associations with Muslims, F(1, 23) = 9.62, p < .01. Moreover, the interaction effect between condition and news consumption was significant when comparing between conditions. The results indicated that high-news-media consumers who were primed with the word terrorist in the news story headline were less likely to associate Muslims with positive hypothetical statements than similar participants in the control condition, F(1, 56) = 6.79, p = .01. Therefore, H2 was supported.

A t-test showed that the Christian participants associated more positive statements with Muslims (m = 0.82, SD = 0.85) than did non-Christians (m = 0.67, SD = 0.80). However, this difference was not statistically significant at two-tailed t(58) = 0.68, p = .50. A second t-test showed that Christians associated more negative statements with Muslims (m = 4.92, SD = 2.85) than did non-Christians (m = 3.00, SD = 2.49). This difference was statistically significant at two-tailed t(58) = 2.60, p = .01.

A general linear model on positive associations with Muslims showed that religiosity had no main effect, F(1, 56) = .08, p = .77, and that the interaction between condition and level of religiosity was not significant, F(1, 56) = 0.15, p = .70. However, a general linear model on negative associations with Muslims showed that religiosity had a significant effect, such that higher levels of religiosity were associated with more negative associations with Muslims, F(1, 56) = 4.68, p = .034.

Conclusions

The data indicate that the Muslim terrorist stereotype does exist, as Islam was the only religion (of the five analyzed) found to be more commonly associated with negative than with positive hypothetical statements. Also, the results show that if primed with the word terrorist in a news story headline, participants who reported consuming more news media associated fewer positive statements with Muslims, which suggests a cultivation effect. Two effects were found across both conditions: participants who self-reported as Christians associated more negative statements with Muslims, and level of religiosity was found to be positively correlated with negativity toward Muslims.

However, participants were more likely to associate Muslims with negative statements, regardless of condition and level of news media consumption. This finding suggests that negativity toward Muslims may be so pervasive in an American setting that any implicit priming effects are mitigated by explicitly held beliefs by participants in both conditions of the experiment.

For more details regarding the study contact

Dr. S. Shyam Sundar by e-mail at sss12@psu.edu or by telephone at (814) 865-2173

>> Return to abstract listings page
>> Return to the main research page

Media Effects Research Lab at College of Communications, Penn State University