LECTURE 7 Flashcards
General structure of the experiment
Experimental design
Design to be used is determined mainly by the nature of
hypothesis.
Experimental design
DESIGN IS BASED ON THE FOLLOWING;
● Number of ___
● Number of ___ needed to make a fair test
● Whether the ___or ___ subjects are used in each of the
treatment conditions
IVs
Treatment conditions
Same or different
DIFFERENT subjects take part in EACH
condition of the experiment.
Between subject design
Draw conclusions by making
comparisons between the behaviors
of different group subjects.
Between subject design
The more the sample resembles the whole population, the
more likely it is that the behavior or performance of the sample
mirrors that of the population.
Selecting and recruiting subject
___procedure is highly desirable because it
increases the external validity of an experiment; rarely done in
practice.
Random sampling
❑ Too __ sample size can lead to incorrect results.
- not representing the whole population
- may decrease the power of IV
❑ Number of subjects needed can depend on the expected amount
of variability among subjects’ scores and effect size
Small
*If individuals in the population are all ___ to one another on the
dependent variable, ___samples sizes are adequate.
*When individuals are quite ___ on the dependent variable, ___ samples
are needed.
Very similar
Smaller
Different
Larger
the statistical estimate of the size or magnitude of
the treatment effect.
Effect size
the ___ the ___, the ___ the subjects needed to determine the
treatment effect; ___ effects may be detected in ___groups.
Larger
Effect size
Fewer
Weaker
Larger
____/___or softwares such as G*Power can be used to
estimate the ___ number of subjects needed for each treatment group,
given the effect size.
Power charts
Table
Minimum
As general rule, at least ___subjects in each treatment group is
advisable; some are more comfortable in __ subjects per
treatment group.
20
30
Simplest experimental design.
Two groups design
1IV; Two treatment conditions are needed; two separate
groups of subjects are formed.
Two groups design
Used when an independent variable must be tested at two
treatment levels or values
Two-independent groups design
every subject has an equal chance of
being placed in any of the treatment conditions
Random assignment
assignment is still random except
for limitations on numbers per group, etc.
Random assignment with constraints
Random ____is essential for ___ validity, whereas
random ___ is critical to ____ validity
Selection
External
Assignment
Internal
Two-independent-groups design can be further classified into
two: ____ group design and
____-groups design
experimental group-control
two-experimental
Consists of an experimental condition and
control condition
Two-Independent-Groups Design: Experimental Group-Control Group Design
we apply a particular
value of our IV to the subjects; the subjects in
this condition are called an ____
Experimental condition
Experimental group
used to determine the value
of DV without an experimental manipulation of
the IV; subjects in this condition are called as a
control group
Control condition
E.g., Effects of music in learning of college
students.
Two-Independent-Groups Design: Experimental Group-Control Group Design
Experimental design wherein there are at least two groups, one
of which does not receive a treatment or intervention, and data
are collected on the outcome measure after the treatment or
intervention
Posttest-Only Control Group Design
EG - ____
CG - ____
R - ____
X - ____ (Independent Variable)
O - _____ (Dependent Variable)
Experimental group
Control group
Randomly assigned participant group
experimental manipulation
Observation/measurement
Subjects are randomly assigned to experimental and control
groups. Both are subjected to an initial (pretest) measurement
of the dependent variable of interest. The experimental group
is administered a treatment. The dependent variable is
measured again for both groups (posttest).
Pretest-Posttest Control Group Design
❑ A design in which two experimental groups are exposed to
different levels of the independent variable.
❑ Two-experimental-groups design becomes more applicable
when studies had established that an IV produces some effect
on a DV.
Two-Independent-Groups Design: Two-Experimental-Groups Design
The influence of music on reading comprehension
Independent Variable (IV) - Music
Levels/Values: Classical Music (X1
) and Rock Music (X2
)
Two-Independent-Groups Design: Two-Experimental-Groups Design
In this design, there are also two groups of subjects, but they
are assigned to groups by matching or equating them on a
possible extraneous subject variable
Two-Matched-Groups Design
To match subjects before the experiment:
1. Subjects will be measured on the extraneous variable
that will be used for matching (e.g., IQ)
2. We separate them into pairs with similar scores
3. We randomly assign one member of each pair to a
treatment condition, then the remaining is placed on
the other
4. We run the experiment
To match subjects after the experiment:
1. We randomize and run the experiment, and we also
measure the extraneous subject variable as well (e.g., IQ)
2. After the experiment, we will use their extraneous variable
scores to match the subjects across the two groups
3. Subjects who cannot be matched are discarded
Two-Matched-Groups Design
3 ways to match subjects across the two groups after experiment:
➔ Precision matching
➔ Range matching
➔ Rank-ordered matching
There are more than two groups of subjects, and each group
is run through a different treatment condition; used when
the amount or degree of the IV is important (e.g., drug
dosage)
Multiple Groups Design
most commonly used multiple groups design; the subjects
are assigned to the different treatment conditions
randomly
Multiple independent groups design
most commonly used multiple groups design; the subjects
are assigned to the different treatment conditions
randomly
Multiple independent groups design
The use of random number table (e.g., online generator) is one
way to randomly assign subjects into multiple treatment
conditions
Multiple Groups Design
To obtain equal numbers of subjects in all treatment
conditions, we can use ____
block randomization
a method for
assigning subjects to treatment conditions in which individuals
are divided into subsets or blocks and then some random
process is used to place individuals from those blocks into the
different conditions
Block randomization
a research design that attempts to consider the influence of pretesting
on subsequent posttest results.
Solomon four group design
