Transcript Document
Chapter 21
Research Design Applications:
Randomized Groups and
Correlated Groups
Simple Randomized Subjects
Design
• The general design paradigm (designated
as Design 20.1) is shown in p.502.
• Research example
• Dolinski and Nawrat (1998): Fear-thenRelief and Compliance.
• They claim that compliance is due to the
reduction of fear and not the fear itself. A
one-way analysis of variance is used for
the study. Table 21.1
Factorial Designs
• Factorial design is the structure of research in which
two or more independent variables are juxtaposed
in order to study their independent and interactive
effects on a dependent variable.
• The simplest factorial design, the 2*2, has three
possibilities: both A and B active; A active, B
attribute (or vice versa); and both A and B attribute.
(The last possibilities, both independent variables
attributes, is the nonexperimental case. As indicated
earlier, however, it is probably not appropriate to
use analysis of variance with nonexperimental
independent variables.)
Factorial Designs with More than
Two Variables
• Each cell must have at least two
participants, and preferably many more. (It
is possible, but not very sensible, to have
only one participant per cell if one can
have more. Of course, there are designs
that have only one participant per cell.
This is covered in Chapter 22.)
Research Examples of Factorial
Designs
• Sigall and Ostrove (1975): Attractiveness
and Crime.
• They asked the question: Ho is the
physical attractiveness of a criminal
defendant related to juridic sentences, and
does the nature of the crime interact with
attractiveness?
• Table 21.2
Research Examples of Factorial
Designs
• Hoyt (1955): Teacher Knowledge and
Pupil Achievement.
• The research question was: What are the
effects on the achievement and attitudes
of pupils if teachers are given knowledge
of the characteristics of their pupils?
• A 3*3*2*2 factorial design with variables
Treatment, Ability, Sex, and School. Figure
21.1 and Table 21.3.
Evaluation of Randomized Subjects Designs
• Randomized subjects designs are all variants
or extensions of Design 20.1, the basic
experimental group-control group design in
which participants are assigned to the
experimental and control groups at random.
• As such they have the strengths of the basic
design, the most important of which is the
randomization feature, and the consequent
ability to assume the preexperimental
approximate equality of the experimental
groups in all possible independent variables.
Evaluation of Randomized Subjects Designs
• Two other strengths of these designs,
springing from the many variations
possible, are flexibility and applicability.
These can be used to help solve many
behavioral research problems, since they
seem to be peculiarly well suited to the
types of design problems that arise from
social scientific and educational problems
and hypotheses.
Evaluation of Randomized Subjects Designs
• There are also weakness. One criticism
has been that randomized subjects
designs do no permit tests of the equality
of group, as do before-after (pretest-post
test) designs. Actually, this is not a valid
criticism for two reasons: (1) with enough
participants and randomization, it can be
assumed that the groups are equal, as we
have seen; an (2) it is possible to check
the groups for equality on variables other
than Y, the dependent variable.
Evaluation of Randomized Subjects Designs
• Another weakness is statistical. One
should have equal numbers of cases in
the cells of factorial designs. It is possible
to work with unequal ns, but it is both
clumsy and a threat to interpretation.
Multiple regression is a better solution for
this problem.
Evaluation of Randomized Subjects Designs
• Compared to matched groups designs,
randomized subjects designs are usually
less precise; that is, the error term is
ordinarily larger, other things being equal.
Correlated Groups
• A basic principle is behind all correlated
groups designs: there is systematic
variance in the dependent variable
measures due to the correlation between
the groups on some variable related to the
dependent variable.
• This correlation and its concomitant
variance can be introduced into the
measures, and the design, in three ways:
Correlated Groups
• 1. use the same units, for example,
participants, in each of the experimental
groups,
• 2. match units on one or more
independent variables that are related to
the dependent variable, and
• 3. use more than one group of units, like
classes or schools, in the design.
The General Paradigm
• The word group should be taken to
indicate set of scores. Then there is no
confusion when repeated trials experiment
is classified as a multigroup design.
• Figure 21.2
• It can be seen that there are two sources
of systematic variance: that due to column,
or treatments, and that due to rows
(individual or unit differences).
Units
• The word unit is deliberately used to
emphasize that units can be persons or
participants, classes, schools, districts,
cities, even nations.
• In Figure 21.3. On the left is a factorial
design and on the right a correlated
groups design, but they look the same!
They are the same, in variance principle.
(The only differences might be numbers of
scores in the cells and statistical
treatment.)
One Group, Repeated Trials
Design
• It was said earlier that the best possible
matching of participants is to match the
participant with himself or herself. One of
the difficulties using this solution
resembles pretest sensitization, which
may produce an interaction between the
pretest and the experimentally
manipulated variable. Another is that
participants mature and learn over time.
One Group, Repeated Trials
Design
• The problem of how individuals learn, or
become unduly sensitized during an
experiment, is difficult to solve. In short,
history, maturation, and sensitization are
possible weakness of repeated trials. The
regression effect can also be a weakness.
• The simplest case of this kind of designs is
one group, Before-After design, Design
19.2 (a).
Two Groups, Experimental GroupControl Group Designs
• This design is described as Design 20.2.
• The most common variant of the group,
experimental group-control group design is
the Before-After, two group design [see
Design 20.3 (b)]
Research Examples of Correlated
Group Designs
• Miller and DiCara: Learning of Autonomic
Functions
• Table 21.4
• The research design is a variant of Design 20.3 (a)
• The difference is that ~X, which in the design
means absence of experimental treatment for the
control group, now means reward for decrease of
urine secretion. The usual analysis of the aftertraining measures of the two group is therefore
altered.
Research Examples of Correlated
Group Designs
• Tipper, Eissengberg, and Weaver (1992):
Effects of Practice on Selective Attention
• A completely within-subjects design. All of
the participants experienced all of the
treatment conditions.
Multigroup Correlated Groups
Designs
• Units Variance
• Until recently, the variances due to
differences between classes, schools,
school systems, and other “natural” units
have not been well controlled or often
used in the analysis of data.
• The educational investigator has to be
alert to these unit differences, as well as to
individual differences.
Multigroup Correlated Groups
Designs
• Factorial Correlated Groups
• Figure 21.5
• Suedfeld and Rank (1976): Revolutionary
and Conceptual Complexity
• Table 21.5
Multigroup Correlated Groups
Designs
• Perrine, Lisle, and Tucker (1995): Offer of Help
and Willingness to Seek Support
• The design was a 3*2*2 factorial design. It
contained one manipulated (active) independent
variable, one measured (attribute) independent
variable and one within-subjects (correlated)
independent variable. Figure 21.6
• It usually referred to as mixed ANOVA when at
least one independent variable is betweensubjects and at least one other is within-subjects.
Analysis of Covariance
• Analysis of covariance is a form of
analysis of variance that tests the
significance of the differences among
means of experimental groups after taking
into account initial differences among the
groups, and the correlation of the initial
measures and the dependent variable
measures. The measure used as a control
variable—the pretest or pertinent
variable—is called a covariate.
Analysis of Covariance
• Clark and Walberg (1968): Massive
Reinforcement and Reading Achievement
• Table 21.7
Research Design and Analysis:
Concluding Remarks
• Four major objectives have dominated the
organization and preparation of Part Six.
• 1. to acquaint the student with the principal designs
of research.
• 2. to convey a sense of the balanced structure of
good research designs, to develop sensitive feeling
for the architecture of design.
• 3. to help the reader understand the logic of
experimental inquiry and the logic of the various
design.
• 4. to help the student understand the relation
between the research design and statistics.