Chapter 4: Experimental Designs information: They give us information about

the effects of each independent variable in

Experimental Design the experiment, called main effects.
- The general structure of an experiment (but
not its specific content). Main Effects
- The purpose of experimental design is to - is the action of a single independent variable
minimize extraneous or uncontrolled in an experiment.
variation, thereby increasing the likelihood - A main effect is simply a change in
that an experiment will produce valid, behavior associated with a change in the
consistent results. value of a single independent variable within
the experiment
3 aspects of experiment that play the biggest part
in determining the experimental design; Interaction
1. The number of Independent Variables. - is the effect of one independent variable
2. The number treatment conditions needed to changes across the levels of another
make fair test of the experimental hypothesis. independent variable; can only be detected
3. Whether the same or different subjects are in factorial designs.
used in each of the treatment conditions. - Higher-Order Interaction- An interaction
effect involving more than two independent
Between Subject Designs variables.
- A design in which different subjects take
part in each condition of the experiment. Within-Subjects Design
- The between-subjects (two groups) design is - A design in which each subject takes part in
conservative. There is no chance that one more than one condition of the experiment;
treatment will continue to contaminate the also called a repeated-measures design.
other, because each person receives only one
treatment (one repletion or five repetitions, Within-Subjects Factorial Design
but not both.) - A factorial design in which subjects receive
all conditions in the experiment.
Experimental Group-Control Group Design
- Experimental Group- Receives the Mixed Designs (Combining Within- and Between-
independent variable (treatment or Subjects)
intervention). - A factorial design that combines within-
- Experimental Condition- The specific subjects and between-subjects factors.
treatment or intervention given to the
experimental group. Order Effects
- Control Group- Does not receive the - Change in subjects’ performance that occurs
independent variable (serves as a baseline). when a condition falls in different positions
- Control Condition- The condition where no in a sequence of treatments.
treatment is given, allowing for comparison.
Practice Effect
Placebo Group - Change in subjects’ performance resulting
- A special type of control group where from practice.
participants receive a fake treatment but
believe it is real. Carryover Effect
- Used to control for psychological effects - The persistence of the effect of a treatment
that may influence results. condition after the condition ends.
Two-Experimental-Groups Design Progressive Error
- Instead of a control group, both groups - Changes in subjects’ responses that are
receive different levels of the independent caused by testing in multiple treatment
variable. conditions; includes order effects, such as
the effects of practice or fatigue.
Factorial Designs
- Designs in which we study two or more Counterbalancing
independent variables at the same time. - A technique for controlling order effects by
- The independent variables in these designs distributing progressive error across the
are called Factors, and each factor will have different treatment conditions of the
two or more values or levels. experiment; may also control carryover
effects.
Between Subject Factorial Designs
- each level of one independent variable is Across-Subjects Counterbalancing
combined with each level of the others to - A technique for controlling progressive error
produce all possible combinations that pools all subjects’ data together to
- The simplest factorial design has only two equalize the effects of progressive error for
factors. each condition.
- The results we get from a factorial
experiment give us two kinds of
 - Complete Counterbalancing- A technique - The hawthorne effect is when subjects of
for controlling progressive error using all an experimental study attempt to change or
possible sequences that can be formed out of improve their behavior simply because it is
the treatment conditions and using each being evaluated or studied.
sequence the same number of times.
- Partial Counterbalancing- A technique for Experimenter Effect
controlling progressive error by using some - The experimenter effect influences the
subset of the available sequences of outcome accidentally by providing
treatment conditions. participants with slight cues as to the
a. Randomized Partial Counterbalancing- experimenter’s expectations.
The simplest partial counterbalancing
procedure in which the experimenter Double-Blind Experiment
randomly selects as many sequences of - It is termed as a “double-blind
treatment conditions as there are subjects for experiment” because neither the
the experiment. experimenter nor the research participant
b. Latin square counterbalancing- A knows which subjects are in which treatment
partial counterbalancing technique in which conditions
a matrix, or square, of sequences is
constructed so that each treatment appears Automation of Experiments
only once in any order position. - Automation of experiments refers to using
technology, software, or machines to run
Subject-by-Subject Counterbalancing experiments with minimal human
- A technique for controlling progressive error intervention. This reduces experimenter
for each individual subject by presenting all bias, ensures standardized procedures, and
treatment conditions more than once. allows efficient data collection.
- Reverse Counterbalancing- A technique
for controlling progressive error for each
individual subject by presenting all
treatment conditions twice, first in one
order, then in the reverse order.
- Block Randomization- A process of
randomization that first creates treatment
blocks containing one random order of the
conditions in the experiment; subjects are
then assigned to fill each successive
treatment block.

Small-N design
- Small-N design test only one or a very few
subjects ; typically, the experimenter
collects baseline data during an initial
control condition, applies the experimental
treatment, then reinstates the original control
condition to verify that changes observed in
behavior were caused by the experimental
intervention.

Baseline
- A measure of behavior as it normally occurs
without the experimental manipulation; a
control condition used to assess the impact
of the experimental condition.

Pitfalls
- A pitfall is a potential issue or flaw that can
negatively impact the validity, reliability, or
ethical integrity of a study.

Demand Characteristics
- Demand characteristics are cues that might
indicate the research objectives to the
participants. These cues can lead
participants to change their behavior or
responses based on what they think the
research is about.

Hawthorne Effect