The Bulletin’s survey about diversity and the language framing immigration is based on best sampling practices and represents, with a high degree of confidence, the attitudes of undergraduates at Emporia State.
The survey was prompted by the controversy that began when Michaela Todd, ASG vice president and senior political science and communications major, made a political Facebook post which referenced “illegal aliens.”
The controversy has lasted three weeks and has prompted student outcry, a call for Todd’s impeachment and an ASG meeting to be canceled due to “safety concerns.”
“I think your poll was very matter of fact,” said Michael Smith, professor of social sciences. “You only asked a few questions and they were pretty straight forward. I think the poll is good in that you didn’t try to do things with it that you didn’t have the data to do...You have to know the limits of your data.”
When addressing the poll’s reliability, the important factors to examine are the margin of error, the number of people sampled and whether the poll had non-leading questions, according to Smith.
The Bulletin’s poll had a sample size of 414 from a total population of 3,605 undergraduate students at ESU. Based on these numbers and the methodology, the survey has a confidence level of 95 percent and a plus or minus 4.5 percentage points margin of error.
This was calculated using surveysystem.com.
“The confidence level...is expressed as a percentage and represents how often the true percentage of the population who would pick an answer lies within the confidence interval,” according surveysystem.com. “Most researchers use the 95 percent confidence level.”
In order to increase the reliability of the responses, The Bulletin randomized the order of the questions, asked non-leading questions and went to randomly selected classes to administer the survey, with the permission of the professors.
“What you did was probably closest to a stratified sample, where you take, in your case, different classes taught by different professors in different disciplines,” Smith said. “The gold standard of samplings is called ‘random sampling’ where every member of the population has an equal chance of getting selected. But random samples are unattainable in the real world, and the stratified sample can work if it is pretty representative of the group.”
The surveys were anonymous and done on a volunteer basis. The polls were given between Nov. 14-28.
No one paid The Bulletin for the survey or funded the materials and hours spent administering the poll.
“Polls are probably more useful for determining the climate of public opinion,” said Michael Smith, professor of social sciences. “They aren’t perfect but you have to remember that lots of disciplines use sampling. You can’t test all of the air in Kansas to determine air quality. It’s impossible. You have to sample. You can’t test all of the water in the Cottonwood river to see if it’s polluted. You have to sample.”
Smith said he felt the poll was representative of campus.
“It’s kind of a double standard for people to accept those things and say ‘Well I don’t believe the result of a public opinion poll because their sampling,’” Smith said. “It’s all about how well the sampling was done and how well the poll was done.”
The Bulletin is not under the purview of the Institutional Review Board, which provides oversight of research that involves human subjects, according to niehs.nih.gov.
“I spoke with a representative from the IRB who stated that the IRB only handles issues that relate to research and that would involve things like (academic) publication,” Smith said. “It does not exempt The Bulletin from following best practices to protect anonymity of subjects and to be mindful of any harm…But you are not under the jurisdiction of the IRB because a poll published in the student newspaper is not considered institutional research.”
Smith said a further step for the poll was to compare the responses from individual ethnic and culture groups, instead of people of color as a whole. However, Smith also said that ESU does not have enough people of color to do this.
“If you had the resources, it would be really neat to take things to the next level with the whole First Amendment issue,” Smith said. “Trying to tease out people...who deplore the statement ‘illegal alien’ but would defend someone’s right to use it. Would you uphold the free speech of someone whose speech offends you?”