The Impact of Air Shows, Fly-overs, Open Houses, and Guest Days
on Public Opinion
Jon Connor, Patricia Huizinga, Peter Kerr
|In the pilot study, a three-question
survey designed to locate attitude change was drafted and tested at the
April 2000 air show hosted by Altus AFB, Oklahoma. Three researchers were
trained in how to give the survey identically, and then collected data
from 10:30 a.m. until 2 p.m. Two-hundred thirty-two people entering the
air show and 141 people exiting the air show were polled on a seven-point
Likert scale. The survey results and exact introduction script and questions
are in Appendix B. For the pilot study, pre-test and post-test subject
numbers varied, as no $5 awards were available. The estimated crowd attendance
was 30,000 people.
During the pilot study, very few air show attendees turned down the opportunity to register their opinions. The questions were found to be comprehensible to 95% of the public without any further explanation. The explanation that did have to occur mostly had to do with a definition of the term "support," as a few air show attendees saw the word to connote finances. Other notable occurrences while conducting the pilot survey were that some people felt themselves disqualified from answering since they were in the military. And some amount of peer pressure may bias the data as couples surveyed simultaneously often had similar or identical answers. This last deficiency would be overcome in the future by having the survey on individual pieces of paper instead of being gathered verbally.
On entering, the mean response for question one was 6.409 while the exiting mean was 6.433. The pre-test for question two has a mean of 6.263 while the post-test has a mean of 6.241. The entering data for question three has a mean of 6.724, while the exiting mean was 6.823. While questions one and three had a higher pre-test mean than post-test mean, a t-test revealed no significant difference between entrance and exiting data for all three questions.
Less than 3% of those attending the airshow, in either the pre-test or post-test survey, had lower than a neutral response to the three questions. On all three questions, over 80% of the answers were 6 and 7 (agree and strongly agree) on the Likert scale.
The pilot study results succeeded in refining the survey process and questions, but did not capture any information variance beyond what can be expected by chance. It is interesting to note that while people exiting more often strongly agreed with questions 1 and 3 than upon entering, there was a drop of people reporting strongly agree to the question of proper use of taxes. This difference may be magnified in future studies.
Another interesting finding is that the data shows an overwhelming amount of support for the military. Even if people did not strongly agree with the use of taxes for air shows, or even have a strong favorable opinion of the military, they still self-reported as being strong supporters of the military (78% entering, 88% exiting). It is conceivable that patriotism is at work, as they see the question of supporting the military a part of their civic duty. There may also be a "door in the face" interaction, as they may have answered lower on the tax question, causing some guilt, and then decided to justify themselves by stating their strong support for the military.
One important finding of the pilot test is that air show attendees already hold very pro-military sentiments. Less than 3% of air show attendees answered lower than neutral on any of the three questions. This hints at air shows being a good way to assist retention, and maintain positive relations with the public, but lowers the possibility of an air show actually changing the opinion of someone with a preconceived negative disposition toward the military.
One important confound to the pilot study may be the fact that the data was collected between 10 a.m. and 2 p.m. This means that those leaving the air show did not consider the air show worth attending all day, and thus the exiting subjects polled may represent the less positive attendees. This confound will be overcome in the future by the implementation of the ticketing and $5 reward, ensuring the same pre-test and post-test sample.