Jackson S.L. Research Methods and Statistics: A Critical Thinking Approach
Подождите немного. Документ загружается.
332
■ ■
CHAPTER 12
the possibility that some other extraneous variable produced the results.
In other words, because behavior changed for each participant right after
the treatment was introduced and because the treatment was introduced at
different times to each participant, we can feel fairly confident that it was
the treatment and not some extraneous variable that caused the behavior
change.
Multiple Baselines Across Behaviors. An alternative multiple-baseline
design uses only one participant and assesses the effects of introducing a
treatment over several behaviors. This design is referred to as a multiple-
baseline design across behaviors. For example, imagine that a teacher
wanted to minimize the number of problem behaviors emitted by a student
during the school day. The teacher might begin by taking baseline meas-
ures on all of the problem behaviors (for example, aggressive behaviors,
talking out of turn, and temper tantrums). The treatment would be intro-
duced first for only aggressive behaviors. Several days after introducing
the treatment for aggressive behaviors, the teacher would introduce the
treatment for talking out of turn, and then several days later for temper
tantrums. By introducing the treatment for different behaviors at differ-
ent times, we can eliminate potential confounds. In other words, if all of
the treatments were introduced at the same time and behavior changed,
we would not know whether the change was due to the treatment or to
some extraneous variable that also changed at the same time. If we see
multiple-baseline design
across behaviors A single-
case design in which measures
are taken at baseline and after
the introduction of the indepen-
dent variable at different times
across multiple behaviors.
multiple-baseline design
across behaviors A single-
case design in which measures
are taken at baseline and after
the introduction of the indepen-
dent variable at different times
across multiple behaviors.
10017_12_ch12_p316-338.indd 332 2/1/08 1:34:14 PM
Text not available due to copyright restrictions
Quasi-Experimental and Single-Case Designs
■ ■
333
a systematic improvement across behaviors when the treatment is intro-
duced at different times, we can feel fairly certain that it was the treatment
that brought about the change.
Multiple Baselines Across Situations. A third way to use the multiple-
baseline design is to assess the introduction of treatment across different
situations—a multiple-baseline design across situations. For example,
Hall and his colleagues (1971) assessed the effectiveness of punishment on a
young boy’s crying, whining, and complaining behavior during school. The
child emitted these behaviors only during reading and math classes each
day. Hall devised a system in which the child was given five slips of colored
paper bearing his name at the beginning of reading and arithmetic periods
each day. One slip of paper was taken away each time he cried, whined, or
complained. As can be seen in Figure 12.5, baseline performance was estab-
lished for the number of cries, whines, and complaints in each class. Then the
treatment was introduced on day 6 in the reading class and on day 11 in the
math class. In both situations, the number of cries, whines, and complaints
declined. Introducing the treatment at different times in the two classes
minimizes the possibility that a confounding variable is responsible for the
behavior change. You can also see that Hall then reversed the treatment and
went back to baseline. Reversal was possible in this situation because the
treatment did not have any carryover effects and because reversing the treat-
ment had no ethical ramifications. The treatment was then reintroduced on
day 21 in both classes. Thus, this design is really a multiple-baseline reversal
design across situations.
multiple-baseline design
across situations A single-
case design in which measures
are taken at baseline and after
the introduction of the indepen-
dent variable at different times
across multiple situations.
multiple-baseline design
across situations A single-
case design in which measures
are taken at baseline and after
the introduction of the indepen-
dent variable at different times
across multiple situations.
10017_12_ch12_p316-338.indd 333 2/1/08 1:34:15 PM
Text not available due to copyright restrictions
334
■ ■
CHAPTER 12
Summary
In this chapter, you were introduced to quasi-experimental designs—a type
of design that falls somewhere between a correlational design and a true
experimental design—and single-case designs. Important concepts related to
quasi-experimental designs include nonmanipulated independent variables
(participant variables), internal validity, and confounds. Quasi-experimental
designs include both single-group designs and nonequivalent control group
designs in addition to developmental designs. Single-case or small-n designs
include reversal designs and multiple-baseline designs. In a reversal design,
the independent variable is introduced and then removed (possibly several
times) in order to assess its effect on the single participant in the study. In a
multiple-baseline design, the independent variable is introduced at different
times across a few participants, behaviors, or situations.
IN REVIEW Single-Case Designs
REVERSAL DESIGNS MULTIPLE-BASELINE DESIGNS
ABA design: Measures taken at baseline, after Across participants: Measures taken at baseline and
introduction of independent variable, after introduction of independent variable at different
and at baseline again times across multiple participants
ABAB design: Measures taken at baseline, after introduction Across behaviors: Measures taken at baseline and after
of independent variable, at baseline again, and after introduction of independent variable at different times
introduction of independent variable again across multiple behaviors
Across situations: Measures taken at baseline and after
introduction of independent variable at different times
across multiple situations
CRITICAL
THINKING
CHECK
12.3
1. Explain why single-case research is considered a variation of a
within-participants design. Why might it also be considered a varia-
tion of a quasi-experimental design?
2. Why is an ABAB design considered more ethical than an
ABA design?
3. How do reversal designs attempt to control for confounds?
4. How do multiple-baseline designs attempt to control for
confounds?
10017_12_ch12_p316-338.indd 334 2/1/08 1:34:15 PM
Quasi-Experimental and Single-Case Designs
■ ■
335
nonmanipulated independent
variable
single-group posttest-only design
single-group pretest/posttest
design
single-group time-series design
nonequivalent control group
posttest-only design
nonequivalent control group
pretest/posttest design
multiple-group time-series design
cross-sectional design
cohort
cohort effect
longitudinal design
sequential design
single-case design
small-n design
reversal design
ABA reversal design
ABAB reversal design
multiple-baseline design
multiple-baseline design across
participants
multiple-baseline design across
behaviors
multiple-baseline design across
situations.
KEY TERMS
(Answers to odd-numbered exercises appear in
Appendix C.)
1. What is the difference between a true experi-
mental design and a quasi-experimental design?
2. A psychology professor is interested in whether
implementing weekly quizzes improves student
learning. She decides to use the weekly quizzes
in one section of her introductory psychology
class and not to use them in another section of the
same course. Which type of quasi-experimental
design do you recommend for this study?
3. If the psychology professor in Exercise 2 had
access to only one section of introductory
psychology, describe how she might use a sin-
gle-group design to assess the effectiveness of
weekly quizzes. Which of the three single-group
designs would you recommend?
4. Identify some possible confounds in each of
the studies you outlined in your answers to
Exercises 2 and 3.
5. What are the similarities and differences
between cross-sectional, longitudinal, and
sequential designs?
6. Give three reasons a researcher might choose to
use a single-case design.
7. Explain what a reversal design is. Identify
advantages and disadvantages to using reversal
designs.
8. How does a multiple-baseline design differ from
a reversal design?
9. When conducting single-case research, why
do we look at graphs of data rather than
statistically analyze the data as with other
designs?
CHAPTER EXERCISES
12.1
1. Gender, religious affiliation, ethnicity, and visual
acuity would all be participant variables.
2. We have slightly more control in a quasi-
experimental study than in a correlational
study because we typically introduce some sort
of treatment. However, because participants
come to us as members of either the control or
the “experimental” group, we cannot conclude
that the independent variable definitely caused
changes in the dependent variable, as we can
when using a true experimental design.
12.2
1. This is a nonequivalent control group posttest-
only design. The participants came to the study
as either smokers or nonsmokers. We then took
posttest measures on them. We cannot conclude
that any observed differences in lung disease
are due to smoking because variables other than
smoking may differ between the two groups.
2. Pretest/posttest designs are an improvement
because they allow us to assess whether the
groups are similar or different before we intro-
duce the treatment and how much they change
CRITICAL THINKING CHECK ANSWERS
10017_12_ch12_p316-338.indd 335 2/1/08 1:34:15 PM
336
■ ■
CHAPTER 12
after the treatment is introduced. By this means,
we can determine whether the groups are in fact
equivalent before the treatment.
12.3
1. Single-case research is considered a varia-
tion of a within-participants design because it
involves studying one or a few participants in
both the control and experimental conditions.
It is also similar to the single-group pretest/
posttest quasi-experimental design in that it
involves taking pretest and posttest measures
on a single participant rather than on a group of
participants.
2. An ABAB design is considered more ethical
than an ABA design because the final condition
involves administering the treatment to the par-
ticipant, rather than leaving the participant with
no treatment (baseline).
3. Reversal designs attempt to control for con-
founds by reversing the baseline and treatment
conditions one or more times to assess the
impact on behavior.
4. Multiple-baseline designs attempt to control for
confounds by introducing the treatment at dif-
fering time intervals to a few different people, to
the same person in different situations, or to the
same person across different behaviors.
Check your knowledge of the content and key terms
in this chapter with a practice quiz and interactive
flashcards at http://academic.cengage.com/
psychology/jackson, or, for step-by-step practice and
information, check out the Statistics and Research
Methods Workshops at http://academic.cengage
.com/psychology/workshops.
WEB RESOURCES
For hands-on experience using the research meth-
ods described in this chapter, see Chapter 9 “Field
Experiments” in Research Methods Laboratory
Manual for Psychology, 2nd ed., by William Langston
(Wadsworth, 2005) or Lab 12 “Single Subject
Designs” in Doing Research: A Lab Manual for
Psychology, by Jane F. Gaultney (Wadsworth, 2007).
LAB RESOURCES
Chapter 12
■
Study Guide
CHAPTER 12 SUMMARY AND REVIEW: QUASI-EXPERIMENTAL AND SINGLE-CASE DESIGNS
In this chapter, you have been introduced to quasi-
experimental designs—a type of design that falls
somewhere between a correlational design and a
true experimental design—and single-case designs.
Important concepts related to quasi-experimental
designs include nonmanipulated independent vari-
ables (participant variables), internal validity, and
confounds. Quasi-experimental designs include both
single-group designs and non-equivalent control
group designs, in addition to the special designs used
by developmental psychologists: cross-sectional,
longitudinal, and sequential designs. Single-case
or small-n designs include reversal designs and
multiple-baseline designs. In a reversal design,
the independent variable is introduced and then
removed (possibly several times) in order to assess
its effect on the single participant in the study. In a
multiple-baseline design, the independent variable
is introduced at different times across a few partici-
pants, behaviors, or situations.
10017_12_ch12_p316-338.indd 336 2/1/08 1:34:16 PM
Quasi-Experimental and Single-Case Designs
■ ■
337
STUDY GUIDE MULTIPLE-CHOICE SELF TEST
Select the single best answer for each of the following
questions. If you have trouble answering any of the
questions, restudy the relevant material.
1. When using a
variable, participants are
assigned to groups.
a. nonmanipulated independent; randomly
b. nonmanipulated independent; not randomly
c. participant; not randomly
d. both b and c
2. Which of the following is a participant variable?
a. ethnicity
b. gender
c. age
d. all of the above
3. Correlational research differs from quasi-
experimental research in that:
a. with correlational research we measure two
variables.
b. with quasi-experimental research there is one
nonmanipulated independent variable and
one measured variable.
c. with quasi-experimental research there is one
manipulated independent variable and one
measured variable.
d. both a and b.
4. Students in one of Mr. Kirk’s classes participate
in new interactive history learning modules.
Students in another class learn history using the
STUDY GUIDE REVIEW EXERCISES
(Answers to exercises appear in Appendix C.)
STUDY GUIDE FILL-IN SELF TEST
Answer the following questions. If you have trouble
answering any of the questions, restudy the relevant
material before going on to the multiple-choice self
test.
1. A
variable is a characteris-
tic inherent in the participants that cannot be
changed.
2. The
design involves giving a
treatment to a single group of participants and
then testing them.
3. A design in which a single group of participants
is measured repeatedly before and after a treat-
ment is a
design.
4. A design in which at least two nonequivalent
groups are given a pretest, then a treatment, and
then a posttest measure is a
design.
5. When participants from different age groups all
serve together in the same experiment, then a
developmental design is being
used.
6. A design in which only one participant is used is
called a
design.
7. A design in which a few participants are studied
is called a
design.
8. A single-case design in which baseline measures
are taken, the independent variable is introduced
and behavior is measured, and the independent
variable is then removed and baseline measures
taken again is a(n)
design.
9. A small-n design in which measures are taken
at baseline and after the introduction of the
independent variable at different times across
multiple participants is a
design.
10017_12_ch12_p316-338.indd 337 2/1/08 1:34:17 PM
338
■ ■
CHAPTER 12
traditional lecture method. After three months, all
students take a test to assess their knowledge of
history. What kind of design did Mr. Kirk use?
a. nonequivalent control group posttest-only
design
b. nonequivalent control group pretest-posttest
design
c. multiple-group time-series design
d. single-group time-series design
5. A problem with nonequivalent control group
designs is that:
a. they are open to many confounds.
b. there is no comparison group.
c. there is no equivalent control group.
d. both a and c.
6. The difference between pretest/posttest designs
and time-series designs is that time-series designs
take
measures.
a. fewer
b. more
c. the same number of
d. more reliable
7. In terms of developmental designs, a
design is being used when the researcher tests the
same participants at difference ages over many
years.
a. matched-participants
b. sequential
c. cross-sectional
d. longitudinal
8. Which of the following is a type of single-case
design?
a. ABA reversal designs
b. multiple baseline across participants
c. time-series design
d. single-group posttest only design
9. The ABA design is generally considered
than the ABAB design because participants
.
a. more desirable; are left with the effects of the
treatment
b. less desirable; are not left with the effects of
the treatment
c. more desirable; are not left with the effects of
the treatment
d. less desirable; are left with the effects of the
treatment
10017_12_ch12_p316-338.indd 338 2/1/08 1:34:17 PM
339
APA Communication
Guidelines
13
CHAPTER
Writing Clearly
Avoiding Grammatical Problems
Reporting Numbers
Citing and Referencing
Citation Style: One Author
Citation Style: Multiple Authors
Reference Style
Typing and Word Processing
Organizing the Paper
Title Page
Abstract
Introduction
Method
Results
Discussion
References
Appendixes and Author Note
Tables and Figures
The Use of Headings
APA Formatting Checklist
Conference Presentations
Oral Presentations
Poster Presentations
Summary
10017_13_ch13_p339-356.indd 339 2/1/08 1:35:33 PM
340
■ ■
CHAPTER 13
Learning Objectives
• Identify and briefly describe the basic components of an APA-format
paper.
• Be familiar with the basic word processing skills necessary to create an
APA-style paper.
I
n this chapter, we will cover the guidelines set forth by the APA for writing
style. APA has very specific writing style guidelines. The guidelines come
from the fifth edition of the Publication Manual of the American Psychological
Association (APA, 2001a). We will discuss general writing style, including clar-
ity and grammar, how to cite others’ work properly and avoid plagiarism, and
basic typing or word processing configurations for your document. In addi-
tion, we will discuss the basic organization of an APA-style paper. Last, we
will briefly discuss guidelines for presenting your research at a conference.
Writing Clearly
The APA style guidelines are intended to facilitate clear writing. First, the APA
recommends an orderly and organized presentation of ideas in the paper.
Toward this end, you should prepare an outline of the paper before you begin
writing. Second, the APA guidelines stress smoothness of expression, or
clear and logical communication. To help meet this goal, provide transitions
from paragraph to paragraph and section to section, do not change topics
suddenly, and make sure you have not omitted something that is necessary
to understand the material being presented. Third, the APA recommends
striving for economy of expression, avoiding wordiness and redundancy.
Following are some examples of wordiness from the APA manual:
Wordy Better
at the present time now
based on the fact that because
the present study this study
In the following examples, again from the APA manual, the italicized words
are redundant and can be omitted:
Six different groups saw
a total of 45 participants
in close proximity
just exactly
has been previously found
The APA manual provides several strategies to improve writing style and
avoid potential problems, including writing from an outline; putting aside
the first draft, then rereading it after a delay; and asking a colleague or peer
to critique the draft for you.
10017_13_ch13_p339-356.indd 340 2/1/08 1:35:34 PM
APA Communication Guidelines
■ ■
341
Avoiding Grammatical Problems
Clarity and smoothness also depend on grammatical correctness. Be sure
to check for subject and verb agreement. If the subject is singular, then the
verb in the sentence must be singular; if the subject is plural, then the verb
should be plural.
Incorrect: Participant apathy as well as performance on the task decrease
with practice.
Correct: Participant apathy as well as performance on the task decreases
with practice.
A pronoun must agree with its antecedent. Pronouns replace nouns
(antecedents). If the antecedent (noun) is singular, then the pronoun
must be singular; if the antecedent is plural, then the pronoun should
be plural.
Incorrect: The participant first entered their four-digit code.
Correct: The participants first entered their four-digit code.
or: The participant first entered his or her four-digit code.
Another common problem in student papers is the misuse of homo-
phones. Homophones are words that sound the same or are pronounced the
same but are spelled differently and have different meanings. For example,
to, too, and two are homophones, as are rite, write, and right, and their, there,
and they’re. Make sure you understand the proper use of each of these
homophones.
Two other errors frequently made by students that resemble homophone
errors are confusing then and than and effect and affect. The word then is an
adverb meaning “at that time”; than is a conjunction meaning “in compari-
son with.” The following examples illustrate correct usage:
Then: I was at work then.
I want to go to the gym first and then go to the store.
Than: She is a better dancer than I.
I would rather go to the game than study for my exam.
The word effect can be a noun or a verb. As a noun, it means “what is pro-
duced by a cause”; as a verb, it means “to bring about or accomplish.”
Effect (noun): The amount of practice had a significant effect on reaction
time.
Effect (verb): I effected a change in the grading policy.
The word affect can also be a noun or a verb. As a noun, it refers to emotion;
as a verb, it means “to act on or to move.”
Affect (noun): The participants in the placebo group maintained a flat
affect.
Affect (verb): The amount of practice affected reaction time.
Other common problems include distinguishing between that and which and
between while and since. That and which are relative pronouns that introduce an
10017_13_ch13_p339-356.indd 341 2/1/08 1:35:35 PM