trial, that you could base a decision on truthful, complete evidence?“ The hypothesis, that the
test group would report less confidence than the control group, was supported (F [1/241] = 8.55,
p<.01). Two inferences follow. Media content was instrumental in influencing a construction of
reality that stands opposed to the conventional view of trial proceedings as relatively pristine.
Additionally, the news stories appear to have been treated as exemplars, generating echo effects
and implicating a representativeness heuristic. A projective evaluation congruent with the
prediction of exemplification theory is evident. That inference is supported by the belief among
test group respondents that behavior chronicled in the newspaper articles is typical of police. Of
145 responses, 118 evaluated the behavior as somewhat to very typical (m=2.06, sd=.65).
H2 was also supported (F [1/248] = 25.38, p<.01). Members of the test group scored lower than
control group members when asked how honest they believed police and other law enforcement
officials were in developing and presenting evidence. Here again, news articles provided
exemplars of the fabrication or distortion of evidence by police, creating a negative reality of law
enforcement credibility and contributing to projective expectations of contaminated judicial
proceedings.
To test H3, an index was constructed consisting of composite scores from two questionnaire
items, each concerned with general impressions regarding fairness of trials and, more broadly,
the United States criminal justice system. The hypothesis, that test group respondents would
view both less positively than would control group respondents, was supported (F [1/251] = 11.88,
p<.01). Here, again, exposure to media reports was instrumental in the construction of a
negative reality.
Post hoc analysis
A post hoc analysis found no statistically significant relationship for either test or control group
between gender, ethnicity, church attendance, family income, or political preference, and
responses to any of the six measures involved in hypothesis testing. Neither were there
statistically significant relationships among any of those variables and whether or not respondents
had been the victim of a crime, the target of law enforcement activity, were acquainted with law
enforcement officers, judges, or attorneys, or had served as jury members.
The inference is that where statistically significant differences were found, they were attributable
to the influence of the experimental intervention, and not demographic or value-based differences
or personal experience. Moreover, among the relative few members of the test group (n=25) who
recalled having previously read newspaper coverage of the same or incidents similar to those to
which they were exposed in the experiment, there was no statistically significant relationship with
any of the six measures used to test hypotheses.
A subset of 32 subjects in the treatment condition was asked to rank the four newspaper articles
that served as stimuli in the experiment (see Methods section) with regard to the impact of each
article on the subject's perceptions of fairness in the United States criminal justice system.
Thirteen subjects ranked the article concerning the Texas case in which multiple defendants were
convicted of drug charges as exerting the greatest impact on a negative evaluation of probable
truthfulness and completeness of evidence presented at trial, r(31)) = .40, p <.05. The story
included a photograph of one of the defendants.
Eleven subjects indicated the fourth article exerted the most influence on perceptions of fairness.
This six-inch item, concerning the murder of a 16-year-old dropped off by police in a rival gang‘s
territory, included a photograph of a deputy district attorney. Here again, subjects indicated they
were less likely, as potential jurors, to have confidence in the truthfulness and completeness of
evidence presented at a trail, r(31) = -.40, p < .05. The other two articles were selected by too few
subjects to permit meaningful analysis (N < 10). The inference is that articles related to law
enforcement malfeasance that contain photography are more like than those that do not to
No comments:
Post a Comment