Sei sulla pagina 1di 4

COMM 150

Assignment 4

1. A week before the 2008 presidential election, with Barack Obama comfortably ahead of John
McCain in most polls, an article in the Washington Post asked, "Could the polls be wrong?" The
article claimed that there was an undercurrent of concern among polling professionals, in part
because of the wide variation in Obama leads: from 2 to 15 percentage points. Assuming that all
of the polls were telephone surveys that used probability sampling, identify and briefly describe
four possible sources of error.
1) Systematic measurement error (social desirability): Respondents can give biased
answer that is favored by the public. Because Obama was potentially the first African American
president, respondents tended to support him to avoid criticism of racism.
2) Coverage error: Researchers could have incomplete sampling frames because
telephone surveys are based telephone directories, which could provide inadequate sampling
frame to the extent that they exclude the poor who cannot afford telephones or the more wealthy
who tend to have unlisted numbers.
3) Non-response error: Difference in the characteristics of those who choose to respond
and those who refuse to or fail to be contacted because of an insufficient address or a wrong
telephone number. They are never at home or they are on vacation. Usually those who have more
spare time and highly educated are more likely to respond, may have similar political views for
presidential elections, and think seriously care about politics.
4) Sampling error: The difference between a population value (American voters) and a
sample of estimate of the value (those who participated telephone surveys) because a sample
rather than a complete census of the population is surveyed. The selected samples may consist of
more young men, make less money, and live alone, which cannot fully represent the entire
American voters.

2. Five common wording problems in constructing survey questions are: (1) lack of clarity or
precision, (2) inappropriate vocabulary, (3) double-barreled questions, (4) loaded word or
leading question, and (5) insensitive wording (See Singleton and Straits). Identify the wording
problem(s) in each of the following questions and then rewrite them to make them more
satisfactory.

a. Fathers should spend just as much time taking care of their children as mothers do. A. Strongly
agree, B. Agree, C. Disagree, D. Strongly disagree.
Wording problem: 4) loaded word or leading question
Rewrite: Who do you think should spend more time caring for children, mothers or
fathers, or equally?
b. Because women are naturally better caregivers, it makes sense for mothers to spend more time
caring for their children than fathers. A. Strongly agree, B. Agree, C. Disagree, D. Strongly
disagree.
Wording problem: 3) double-barreled questions
Rewrite: Women are naturally better caregivers. A. Strongly agree, B. Agree, C.
Disagree, D. Strongly disagree.
Mothers should spend more time caring for their children than fathers. A. Strongly agree,
B. Agree, C. Disagree, D. Strongly disagree.
c. What is your annual personal income? A. $0-$50, 000, B. $50,000-$300,000, C. $300,000 or
more.
Wording problem: 5) insensitive wording
Rewrite: What is your annual personal income? A. Below $50, 000, B.
$50,000-$300,000, C. $300,000 or more
d. Do you think the man should initiate and pay for the first date?
Wording problem: 3) double-barreled questions
Rewrite: Do you think the man should both initiate and pay for the first date?
e. Is the leadership in your family matriarchal, patriarchal, or egalitarian?
Wording problem: 2) inappropriate vocabulary
Rewrite: In your family, does your mother, or your father, or do both of them make major
decisions?

3. Qualitative Analysis: Field Observations

Steps: 1) By yourself, find a location where you are unrecognizable and that has a lot of
interesting people/events to observe (not a problem in L.A.)

E.g.: Cafes, bars, museums, concert venues, tourist spots

2) Turn off your phones and disconnect the internet from your laptops, if applicable. Just enjoy
your surroundings and take in the eclectic atmosphere.

3) Stay at the site for at least 20 minutes.

4) Be open and receptive to your environment.

5) Jot down notes (preferably with a pen and paper to avoid distraction)

6) Turn notes into a 1 page (double spaced) site observation. Be a detective. Explain as many
details as possible about the events, people, etc. that surround your site (macro and micro level
contexts).
Location: Dining Hall

Time: 6:30 pm to 8 pm

Yesterday night I went to Bruin Plate dining hall for dinner. Most people are UCLA

students. Because it was the most popular time for dining, most of the tables were full. The

environment was really loud. Most people sat down in groups ranging from 2 to 6 people, and

those who ate alone almost never looked up and saw the environment around; instead, they only

looked at their phones and tried to detach themselves from the surrounding noisy environment. I

observed five different long tables, which included 50 people, in one and a half hour. There were

three main observations. First, two people (came in a pair) stayed the longest for dinner. There

were 4 pairs total, with two mixed gender, one male pair, and one female pair. There was no

significant difference among these groups. They mostly did not grab the second round of dishes

and were highly engaged with their conversations. They rarely looked at their phones, and spent

most of their time looking at their conversational partner’s eyes. The second observation was in

groups of three or four, students were still engaged with the conversations, but after 10 minutes

of talking, there would always be one person who started to look at his or her phone. One

interesting point is when there is one person looks at the phone, there will always be one or more

people do the same thing. This phenomenon becomes more frequent as the group size becomes

larger. The third observation was when the group reaches to 5 or more people, the frequency of

checking phones starts to decline. The group talk is divided into several small talks, and the

frequency of checking phones goes back to those sitting in a pair or three people. Overall, people

sitting in pairs for dinner look at their phones least frequently regardless of the gender dynamics.

7) Can you think about a hypothesis that you may generate from this observation? How could

you test this?


Hypothesis: The frequency of college students checking cell phones increases as the

group size increases until the group size reaches 5 people because as group size increases, the

benefit of contributing to conversations decrease, and when the group size is large enough to

have various small talks, the benefit of contributing to conversations increase again.

Testing method: observe groups of students in dining hall in the size of 2 to 5. Measure

the duration of time from they sit at the table to the moment they leave the dining hall and the

total times they check their phones.

Potrebbero piacerti anche