Guide to using the learning objectives
1.
Discuss2three
objectives you must meet to conduct a successful survey design. Then, provide3
examples of how a survey might fail to meet these objectives.
2.
List1five
questions researchers should ask before doing survey research.
3.
Generate5
a rationale for the following statement, "good research starts with a good
hypothesis." Describe2 two steps you can take to increase the
chances that your questions relate to your hypothesis.
4.
Provide3a
rationale for the statement: "If you want to know why people do
what they do or think what they think, do not
use a survey design."
5.
There
are four general categories of reasons why people's self-reports may be inaccurate.
a.
Outline3 each of these categories
of reasons.
b. Produce5 an example of a question
that falls into each category.
6.
Examine4each
of the following as they relate to the question, "if participants know, will
they tell?"
a.
social desirability bias
b. observing demand characteristics
c.
following response sets.
7.
Explain2
how nonresponse bias can hurt the external validity of
survey research.
8.
Produce5at
least one advantage and one disadvantage of each of the following types of questionnaires:
a.
self-administered questionnaire
b. investigator-administered
questionnaire
c.
psychological tests.
9.
List1
three advantages and three disadvantages of conducting a telephone
survey.
10.
Generate5
a sample script (at least 5 questions) for a telephone survey asking
about concerns about terrorism and amount of television watched.
11.
For the set of questions you devised in the previous objective (or for any other
questionnaire you have devised), use at least three of the seven tips
listed on page 290 to
propose5 a way to make your survey instrument more valid by making it
more like a psychological test.
12.
Explain2
what is meant by a fixed-alternative
question.
13.
Assume
that you want to ask a question about political party affiliation. Create5 a question about political
party affiliation for each of the following formats:
a.
nominal-dichotomous
b. Likert-type.
14.
Describe2
what is meant by open-ended question.
a.
Describe2 the main advantages
of open-ended questions
b. Describe2 the main disadvantages
of open-ended questions.
15.
Discriminate4
between structured, semistructured, and unstructured
interviews.
16.
Generate5
a list of nine mistakes people make when writing interview questions.
a.
Choose1 any four of these
mistakes and construct3 an example question that makes each mistake.
b. Modify3 each question you
wrote so that it no longer makes the mistake.
17.
Explain2
why sequencing of questions matters.
a.
List5 five rules for sequencing
questions
b. Compose5 a list of five questions
and prioritize6 them (i.e. put them in the order they should be
asked) according to the five rules you listed in 17a.
18.
Define1random
sampling. Explain2 why researchers like to use random
sampling.
19.
Suppose
that your population is a school of 10,000 students and that you can obtain a random
sample of that population. You want to know what percentage of studentsat that
school smoke. Use3 Table 8-4 (page 310) to determine how large your
sample would need to be if you wanted to be 95% confident that the percentage
of people in your sample who claimed to smoke would be within 3% of the
percentage of people who would claim to smoke if you surveyed all
10,000students.
20.
Use
Table 8-4 to decide6 how large your population would have to be before
you would use random sampling rather than surveying the entire population.
Using Table 8-4, justify6 your decision.
21.
Contrast4
each of the following sampling methods with random sampling.
a.
proportionate stratified random
sampling
b. convenience sampling
c.
quota sampling
22.
Rank6
the four sampling methods discussed in text (random sampling,
proportionate stratified random sampling, convenience sampling, and quota
sampling) in terms of their ability to yield a representative sample. Justify6
your rankings.
23.
Discuss2how
ethical issues should affect how you should plan and conduct survey research.
24.
Explain2
the difference between interval and nominal
data.
25.
Imagine
that your data consists of participants' answers to the following three questions:
1. Are you male or female?
2. How many hours do you study per
week?
3. Do you like college?
Explain2 which of those three questions is/are interval
and which is/are nominal.
26.
Explain2
why you should not calculate a mean on either nominal or ordinal data.
27.
Imagine
that you asked a sample of 10 people to answer the questions listed in objective
25. Their responses are listed
below.
|
Gender |
Hours
studied |
Like
college |
|
male |
18 |
no |
|
female |
21 |
no |
|
male |
30 |
yes |
|
female |
14 |
yes |
|
female |
29 |
yes |
|
male |
16 |
yes |
|
male |
14 |
no |
|
female |
14 |
no |
|
male |
22 |
no |
|
female |
27 |
yes |
a.
Produce5 a table comparing
the mean number of hours men study to the mean number of hours women study.
b. Compose5 a table that displays
the relationship between two predictors--gender and liking of college--on hours
studied.
28.
Using
the concepts sample and population, explain2
the difference between descriptive statistics and inferential
statistics.
29.
Describe2two
main reasons for using inferential statistics.
30.
Using
the table in objective 27, we calculated that the men in the sample claimed to study
an average of 20 hours and that the standard error of the mean for men's
studying was approximately 2.24. Construct3 a 95%confidence interval
for the mean number of hours that men study. Explain2 what this 95%
confidence interval means.
31.
Referring
to the survey described in objective 25 (results from that survey are displayed
in objective 27),
a.
Name1 the statistical test
you would use to determine whether women studied more than men. Defend4 your
use of that test.
b. Name1 the statistical test
you would use to examine the relationship between gender, liking college, and
hours studied. Defend4 your use of that test.
c.
Finally, name1 the test
you would use to determine whether women liked college more than men. Defend4
your use of that test.
32.
Explain2
why survey results may be susceptible to Type I errors ("false alarms").
Discuss2 one step you could take to reduce your false alarm
risk.