Sie sind auf Seite 1von 3

Analyzing

Inductive Arguments
A public opinion survey is characterized as an inductive generalization. With
inductive generalizations you are deriving a conclusion about a population from a
consideration of a sample (Moore and Parker 504). According to Critical Thinking
10th Edition, a reputable public opinion survey usually involves between 1,000
and 1,500 in the sample, as well as an elevated confidence level typically at 95
percent, low error margin that is determined by the sample size, and random
selection process to avoid biased results (353-354). I will begin with the analysis of
the stronger survey first and end with a weaker public opinion survey. The main
variance between the two surveys was the method the participants where selected
as well as the sample size.
For the strong survey I selected a survey that examined whether Americans
Attitudes Toward Abortion [is] Unchanged (Saad). The property in question is
whether Americans attitudes towards abortion have changed. Those who
underwent the survey answered three basic questions, [the] morality of abortion:
whether it is morally acceptable or morally wrong, abortion position: whether the
individual was pro-choice or pro-life, and the legality of abortion: whether it should
be legal under any circumstances, legal only under certain circumstances, or illegal
in all circumstances (Saad). The sample size was 1,025 adults (aged 18+) and the
survey took place May 4-8 2016 (Saad). According to our textbook Critical Thinking
10th Edition, a sample size of a 1000+ is big enough to statistically stand for any
population size thus allowing the survey to avoid hastiness (354). The sample
population was taken from individuals residing within the 50 U.S. states as well as
the District of Colombia (Saad). The survey was a telephone survey that includes
60% cellphone respondents and 40% landline respondents (Saad). The numbers
that where dialed where reached by a random-digital-dial system ("How Does the
Gallup Poll Social Series Work?"). The RDD (random digital dial) method was used
as a form of random selection to ensure that those taking part in the survey had an
equal chance to be dialed. According to "How Does the Gallup Poll Social Series
Work? Gallup chooses landline respondents at random within each household
based on which member had the next birthday. This is another factor of the
random selection because it was not a guarantee that the person answering the
phone would be the participant. The RDD allows for a less biased result as it equally
allows members of a population the chance to be selected.
I strongly believe the sample had little to no chance of a super biased
contamination. According to "How Does the Gallup Poll Social Series Work?" Gallup
weights its final samples to match the U.S. population according to gender, age, race,
Hispanic ethnicity, education, region, and population density. The most current U.S.
Census is utilized to determine a population density ("How Does the Gallup Poll
Social Series Work?"). This survey along with every other Gallup Poll Social Series
(GPSS) is weighted to correct for unequal selection probability, nonresponse, and
double coverage of landline and cellphone users in two sampling frames ("How
Does the Gallup Poll Social Series Work?"). The survey was not self-selected where
the results would have highly concluded with an over representation of one view
and an under representation of another thus leading to a biased conclusion. The
target population was for individuals who are 18 and older. The poll provided a

large enough sample and had an error margin of plus or minus 4 at the 95 percent
confidence level (Saad). The survey was strong and I would assign it a solid 8. I am
giving the survey an 8 and not a solid 10 because of the error margin. The error
margin signifies the researchers (those conducting the survey) expectation of failure
in representing the target population. The survey is not perfect but it is a reminder
that we are all human and make mistakes, even if theyre extremely small.
The weaker survey I chose was on opinions on Proposition 57. The property
in question is if the election was being held today, how would you vote on
Proposition 57?(DiCamillo). The participants where given three response choices
yes, no, or undecided (DiCamillo). Proposition 57 allows parole consideration for
nonviolent felons. Authorizes sentence credits for rehabilitation, good behavior and
education. Provides juvenile court judge decides whether juveniles will be
prosecuted as adult (DiCamillo). The cost of this proposition would cost millions
annually. The sample size was 484 participants, which is pretty small and not large
enough to avoid hastiness. In order for the sample size to avoid hastiness it has to be
1,000+ participants. Those who participated where registered voters (who are likely
to vote) in California (DiCamillo). The survey took place September 7-13 2016.
Emails where sent to eligible individuals as an invitation to take part of the survey.
In the email was a link for the recipients to click on that would redirect them to the
survey. The website does not provide the exact amount of emails that where sent
out but only 484 took the survey. The researches, selected voters using a
proprietary sampling technology frame that establishes interlocking targets, so that
the characteristics of the voters selected approximate the demographic and regional
profile of the overall California registered voter population (DiCamillo). There was
a way for bias to creep into the sample. This survey is basically self-selected since
those who received the email had to be motivated on the topic to be part of the
survey. Those who are uninterested or do not hold a strong opinion on the matter
will most likely delete the email and will not take part of the survey. Self-selection
surveys over-represent people who want to be in the sample and under represent
people who dont have strong enough feelings on the issues to respond or who dont
have the time to go to the trouble (Moore & Parker 378). This survey is also a hasty
generalization because the sample size of 484 is not a large enough sample to
properly represent members of a population. Thus leading to results that do not
accurately portray the viewpoints of Californians on Proposition 57. The target
population was registered voters (18 and older) who are likely to vote in California.
I would assign this poll a solid 4. The sample size is too small to properly represent
various opinions/viewpoints from different populations. Also the survey was self-
selected leading to biased results. The results provided do not properly represent
Californians viewpoint on Proposition 57 because the 484 who participated did so
by their own free will leading to an over representation of a viewpoint.
I began with my stronger survey and ended my analysis with my weaker
survey. Selection and sample size plays a major key in creating weak and strong
public opinion polls. The objective is to portray members of a population equally
and this is accomplished by using a successful random selection method, having a
large enough sample (at least 1,000-1,500), a confidence level that is high (mainly
95 percent), and a low error margin determined by the sample (Moore and Parker

353-354). A main difference between the two surveys was the way the respondents
where selected as well as the sample size.

Works Cited
DiCamillo, Mark. "New Field/IGS Poll on Election Issues." Home. Field Research

Corporation, 27 Sept. 2016. Web. 02 Oct. 2016.
"How Does the Gallup Poll Social Series Work?" Gallup.com. N.p., 29 Sept. 2016. Web.

01 Oct. 2016.
Moore, Brooke. Richard Parker. Critical Thinking 10th Edition. New York: McGraw-

Hill, 2012. Print.
Saad, Lydia. "Americans' Attitudes Toward Abortion Unchanged." Gallup.com. Gallup

Inc., 25 May 2016. Web. 01 Oct. 2016.

Das könnte Ihnen auch gefallen