Factorial Design
1. Introduction:
The factorial design is generally the most efficient and valuable design in
psychological research. It is commonly observed that in many behavioral
researches, the investigator is faced with a problem in which he is required to
manipulate two or more than independent variables simultaneously. The
experimental design suited in such a situation is technically known as the
factorial research design. Factorial designs provide a more comprehensive
understanding of how different factors influence the dependent variable and
allow researchers to explore complex relationships in a controlled
experimental setting.
2. Definition:
According to Kirk (1968), A factorial experiment permits an investigator to
evaluate the combined effect of two or more treatments in single experiment.
According to Mc Guigan (1983), One possible design for studying two or
independent variables in a single experiment is the factorial design. A
complete factorial design is one in which all possible combinations of the
selected values of each of the independent variables are used.
According to Chapline (1975), factorial design is an experimental series in
which the effects of several experimental variables are assessed in turn.
According to Kerlinger (2000), 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.
According to Carl Smith, Ellesworth & Aranson (1976), Factorial design is
a design in which two or more independent variables are tested in such a
way that every level of one variable occurs with every level of the other (p)
each different combination of levels defining a condition.
In factorial design, to control an extraneous variable, it is changed into an
independent variable. In factorial design experimenter studies the effect of at
least two independent variables simultaneously and each IV has at least two
levels. Thus the smallest random factorial design is 2×2. In this, the initial
selection of subjects is done at random from the population. After dividing
the subjects into different experimental groups, treatment X is given. Then
investigator studies the effect on dependent variable.
For a design to be called factorial, different subgroups or subjects
must serve under every possible combination of the independent variables.
As far as possible an equal number of subjects in all subgroups is preferred,
although this is not an essential condition for a factorial design. An equal
number of subjects in all subgroups is preferred because it facilitates
statistical computation. The factorial design enables the experimenter to
study the independent as well as the interactive effect of the two or more
independent variables.
3. The Concept of Interaction Effect:
Interaction is one of the most important concepts in research. An interaction
exists between two independent variables if the dependent variable value
that results from one independent variable is determined by the specific
value assumed by the other independent variable. Interactions, which can
only be studied with factorial designs, are of great importance to psychology.
They help us to understand complex behavior, since responses are not simply
determined by one independent variable—rather, behavior is determined by
a complex of stimuli that intricately interact.
They can be used to systematically explore the reasons for conflicting results
in previous experiments by systematically varying a previous extraneous
variable that assumed different values in the two conflicting experiments (Mc
Guigan,F.J.,1983).
4. Types of Factorial Designs: Factorial designs can be of two types:
(i) simple factorial designs and (ii) complex factorial designs.
i. Simple Factorial Designs: In case of simple factorial designs, we
consider the effects of varying two factors on the dependent variable. It is
also termed as a ‘two-factor-factorial design’. Ex.- 2 × 2 , 3 × 4 or 5 × 3
factorial design etc.
Here's a simple example of a 2x2 factorial design:
Research Problem: To investigate the effects of teaching method and class
size on student performance in mathematics.
Independent Variables:
1. Teaching Method (Two levels: Traditional, Interactive)
2. Class Size (Two levels: Small, Large)
In this design the extraneous variable to be controlled by homogeneity is
called the control variable and the independent variable, which is
manipulated, is called the experimental variable. Then there are two
treatments of the experimental variable and two levels of the control
variable. As such there are four cells into which the sample is divided. Each
of the four combinations would provide one treatment or experimental
condition. Subjects are assigned at random to each treatment in the same
manner as in a randomized group design.
Contingency Table
IV1 (Teaching Method)
Traditional Interactive
Mean
IV2 (Class Size)
Small Small
Small Traditional
(1)
Interactive
(2)
-
Large Large
Large Traditional
(3)
Interactive
(4)
-
Mean - - -
Fig.-1: 2x2 Factorial Designs
Main Effects:
1. Teaching Method Main Effect: This assesses the overall impact of teaching
method on student performance, regardless of class size. It compares the
average performance across levels of teaching method.
2. Class Size Main Effect: This assesses the overall impact of class size on
student performance, regardless of teaching method. It compares the
average performance across levels of class size.
Interaction Effect:
It examines whether the effect of teaching method on student performance
depends on the class size and vice versa.
ii. Complex Factorial Designs: A design which considers three or more
independent variables simultaneously is called a complex factorial design. It
is known as ‘multi factor-factorial design.’ In case of three factors with one
experimental variable having two treatments and two control variables, each
one of which having two levels, the design used will be termed 2 × 2 × 2
complex factorial design which will contain a total of eight cells as shown
below in Fig.- 2
Experimental Variable
Treatment A Treatment B
Control Control Control Control
Variable Variable Variable Variable
2 2 2 2
Level I Level II Level I Level II
Level I Cell 1 Cell Cell 5 Cell 7
Control 3
Variable 1
Level Cell 2 Cell Cell 6 Cell 8
II 4
Fig.-2: 2 × 2 × 2 Complex Factorial Design
5. Statistical Analysis of Factorial Designs:
For the statistical analysis of factorial designs, an analysis of variance and
the F-test are used: the total variance is analyzed into among- and within-
groups components.
6. Merits:
i. Efficient Exploration of Variables: Factorial designs allow researchers to
examine the effects of multiple independent variables simultaneously. This
efficiency is especially beneficial when studying complex phenomena
influenced by various factors.
ii. Interaction Effects: Factorial designs facilitate the exploration of interaction
effects, revealing how the influence of one variable may depend on the level
of another. This nuanced understanding is crucial for capturing real-world
complexity.
iii. Resource Efficiency: Conducting a single factorial experiment can be more
resource-efficient than conducting multiple experiments to examine each
variable independently. This is particularly advantageous in terms of time,
cost, and participant recruitment.
iv. Generalizability: By considering multiple factors, factorial designs enhance
the generalizability of findings. Researchers can investigate whether
relationships hold across different levels of other variables, contributing to a
more robust understanding of the phenomenon.
v. Statistical Power: Factorial designs can increase statistical power by
detecting not only main effects but also interaction effects. This contributes
to a more comprehensive analysis and interpretation of the results.
vi. Comprehensive Understanding: Main effects in factorial designs provide
insights into the independent impact of each variable, while interaction
effects reveal the joint influence. This comprehensive understanding enriches
the interpretation of research outcomes.
vii. Real-world Applicability: Many real-world situations involve the
simultaneous influence of multiple factors. Factorial designs allow researchers
to study these situations in a controlled setting, providing insights that are
applicable to complex, multifaceted scenarios.
viii. Flexibility: Factorial designs are flexible and adaptable to different research
questions and contexts. Researchers can expand the design to include
additional factors or levels as needed.
ix. Reduction of Error Variance: By controlling for multiple variables
simultaneously, factorial designs can help reduce error variance, leading to
more accurate and reliable findings.
7. Demerits:
i. Increased Complexity: As the number of factors and levels increases, the
complexity of the design grows. Managing and interpreting data from
factorial designs with multiple factors can become challenging, especially for
researchers and participants.
ii. Increased Resource Requirements: Conducting experiments with multiple
factors and levels may demand more resources in terms of time, money, and
personnel. This can be a limitation, particularly in research settings with
constrained resources.
iii. Potential for Interaction Complexity: While interaction effects provide
valuable insights, they can also introduce complexity in interpretation.
Understanding and explaining complex interactions may require advanced
statistical methods and expertise.
iv. Potential for Confounding: In factorial designs, especially with a large
number of factors, there is an increased risk of confounding variables.
Confounding occurs when the effects of one variable are mixed with the
effects of another, making it challenging to attribute observed effects to
specific factors.
v. Analysis Challenges: Analyzing data from factorial designs can be intricate,
especially when dealing with multiple factors and interactions. Sophisticated
statistical techniques may be required, and misinterpretation of results is a
risk.
8. Conclusion:
While factorial research designs offer numerous advantages, researchers
must carefully consider the potential complexities, resource requirements,
and analysis challenges associated with their use. Factorial Designs is very
powerful tool but it is not applicable in every condition. Considering its
complexities it should be used by an experienced researcher.
9. Reference:
Kerlinger, F.N. & Lee,H. (2000) Foundation of behavioral research (4th ed)
Harcourt College Publishers, New York.
Kothari C.R. (2012) Research methodology: methods and techniques (2nd
ed) Wiley eastern Ltd. New Delhi
Mc Guigan,F.J.,(1983) Experimental psychology: Methods of Research(4th
ed) Prentice-Hall, Inc., Englewood Cliffs, New Jersey
Singh, A. K. (2012) Research methods in psychology, sociology and
education (10 ed) Motilal Banarsidas .
INDEX:
1. Introduction
2. Definition
3. The Concept of Interaction Effect
4. Types of Factorial Designs
5. Statistical Analysis of Factorial Designs
6. Merits
7. Demerits
8. Conclusion
9. Reference
