|
|
- November 6, 2006
Students move from classroom to polling in political science course
- For two hours just one week before the mid-term elections, a group of
Iowa State students spent their evening talking about politics.
And even though these students tend to be more interested and involved in
political campaigns than most, the conversations were definitely one-sided.
The students in "Campaigns and Elections," a course taught by James Hutter,
associate professor of political science, polled a random sample of registered
voters in a 4th Iowa Congressional District.
Using the facilities at the Institute for Social and Behavioral Research
(ISBR) at the ISU Research Park, the students conducted the "real"
poll. Prior to getting on the phones, the students had to take and pass
the human subjects review test and were instructed in polling procedures
by ISBR staff members.
The group utilized a survey the Department of Political Science used in
2003 shortly before the 2004 precinct caucuses. Significant changes were
made in the new questionnaire to highlight the issues and races this year,
including the Iowa governor's race and the race for the U.S. Congress in
the 4th District.
"The polling experience was exactly what we expected," Hutter
said. "Most calls were not completed for a variety of reasons - no
answer, answering machines, and some refusals. That is an important lesson
in itself."
But students did complete 45 interviews with eight students on the phones
at any one time.
"Talking to real potential voters on the phones and recording their
answers through a computer assisted program was beneficial to the students,"
Hutter said. "The students were very good, very professional telephone
interviewers."
Although they analyzed their results, the 45 completed calls fell far short
of the number needed for a scientifically reliable poll.
"We completed just 45 interviews in two hours so the results cannot
be treated as reliable," Hutter said. "While that is not bad it
showed the students how difficult it is to complete the 800 to 1000 (surveys)
needed for most polls."
|
|