Independent Group Design

Independent Group Design
Cont….
Block Randomization
 Block randomization balances subject characteristics and

potential confoundings that occur during the time in which
the experiment is conducted, and it creates groups of equal
size.
One block of Random order of
conditions conditions
ABCDE CAE B D
Block Randomization
 In block randomization, we assign subjects to conditions
one block at a time.
Advantages of Block Randomization
 Produces groups of equal size.

 Controls for time-related variables; every condition is
tested in each block, variables are balanced across the
conditions of the experiment.
 Controls for time-related variables; such as changes in
experimenters.
 Controls for time-related variables; populations from
which subjects are drawn (spring, fall semester)
Internal Validity
An experiment has internal validity when we are able to state
confidently that the independent variable caused differences
between groups on the dependent variable (i.e., a causal
inference).
Threats to Internal Validity
 Randomly assigning intact groups to different conditions of the
independent variable creates a potential confounding due to
preexisting differences among participants in the intact groups.
 Block randomization increases internal validity by balancing
extraneous variables across conditions of the independent variable.
 Selective subject loss, but not mechanical subject loss, threatens
the internal validity of an experiment.
Mechanical Subject Loss
Mechanical subject loss occurs when

 a subject fails to complete the experiment because of an equipment
failure
 If a computer crashes,
 if the experimenter reads the wrong set of instructions,
 if someone interrupts an experimental session.
Mechanical loss is a less critical problem than selective subject loss

because the loss is unlikely to be related to any characteristic of the
subject.
Selective Subject Loss
Subjective Subject Loss Occurs;

 When subjects are lost differentially across the conditions of the
experiment
 When some characteristic of the subject is responsible for the loss
 When this subject characteristic is related to the dependent variable
used to assess the outcome of the study.
How to handle Selective Subject Loss
 One alternative is to administer a pretest and screen out subjects

who are likely to be lost.
 Researchers can give all subjects a pretest. Then, if a subject is lost
from the experimental group, a subject with a comparable pretest
score can be dropped from the control group
 Demand characteristics refer to the cues and other
information that participants use to guide their behavior in
a psychological study.
 Experimenter effect; Potential biases that arise due to the
expectations of the experimenters.
 Placebo control groups are used to control for the

problem of demand characteristics.
 Double-blind experiments control both demand
characteristics and experimenter effects.
ESTABLISHING THE EXTERNAL VALIDITY OF
EXPERIMENTAL FINDINGS
The findings of an experiment have external validity when they can

be applied to other individuals, settings, and conditions beyond the
scope of the specific experiment.
 In some investigations (e.g., theory-testing), researchers may choose
to emphasize internal validity over external validity; other
researchers may choose to increase external validity using sampling
or replication.
 Mook (1983) argued that, An experiment is often done to determine
whether subjects can be induced to behave in a certain way.
ESTABLISHING THE EXTERNAL
VALIDITY OF EXPERIMENTAL FINDINGS
 Conducting field experiments is one way that researchers can
increase the external validity of their research in real-world settings.
 Field experiments; For example, to investigate people’s perceptions
of risks, participants in two field experiments were asked to answer
questions about risks during the 2009 H1N1 flu pandemic (Lee,
Schwarz, Taubman, & Hou, 2010). (university campus, shopping malls and
near downtown businesses)
 Partial replication is a useful method for establishing the external
validity of research findings.(metropolitan university hospital and in a small
rural community clinic)
ESTABLISHING THE EXTERNAL
VALIDITY OF EXPERIMENTAL FINDINGS
 Researchers often seek to generalize results about conceptual
relationships among variables rather than specific conditions,
manipulations, settings, and samples.
 Insults increase aggressive behavior (Anderson and Bushman, 1997).
Consider a study with 5-year-old children to determine if a specific
insult (“pooh-pooh-head”) induces anger and aggression. We could
do a replication to see if the same insult produces the same result
with 35-year-old adults.
 What about when results in the lab and the real world disagree?
Anderson and Bushman (1997) argue that these discrepancies,
rather than evidence for the weakness of either method, should be
used to help us refine our theories about aggression.
MATCHED GROUPS DESIGN
 A matched groups design may be used to create comparable groups

when there are too few subjects available for random assignment to
work effectively.
 Matching subjects on the dependent variable (as a pretest) is the
best approach for creating matched groups, but scores on any
matching variable must correlate with the dependent variable.
 After subjects are matched on the matching variable, they should
then be randomly assigned to the conditions of the independent
variable.
 Suppose researchers wish to compare two types of postnatal
care for premature infants and it is not possible to give both
types of care to each infant. Six pairs of premature infants
could be matched on their initial body weight.
 The two groups of premature infants may not be
comparable in their general health or in their degree of
parental attachment.
 Matching and than Random Assignment
NATURAL GROUPS DESIGN
 Experiments involving independent variables whose levels are

selected, not manipulated,—like individual differences variables,
gender, age—are called natural groups designs.
 The natural groups design represents a type of correlational research
in which researchers look for covariations between natural groups
variables and dependent variables.
 Causal inferences cannot be made regarding the effects of natural
groups variables because plausible alternative explanations for
group differences exist.
If we were interested in the relationship between divorce and
emotional disorders
 Does divorce precede the emotional disorder, or does the
emotional disorder precede the divorce?

Independent Group Design

Uploaded by

Copyright:

Available Formats

Independent Group Design

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Independent Group Design

Uploaded by

Copyright:

Available Formats

Independent Group Design

 Block randomization balances subject characteristics and

 Produces groups of equal size.

Mechanical subject loss occurs when

Mechanical loss is a less critical problem than selective subject loss

Subjective Subject Loss Occurs;

 One alternative is to administer a pretest and screen out subjects

 Placebo control groups are used to control for the

The findings of an experiment have external validity when they can

 A matched groups design may be used to create comparable groups

 Experiments involving independent variables whose levels are

You might also like