Definition

A factorial design is one in which two or more variable or factors are employed in such a way that
all the possible combinations of selected values of each variable are used (Mcburney & White,
2007). According to Singh (1998), Factorial design is a design in which selected values of two or more
independent variables are manipulated in all possible combinations so that their independent as
well as interactive effect upon the dependent variable may be studied.

On the basis of the above definition it can be said that the factorial design is one in which two or
more independent variables are manipulated in all possible combinations and thus the factorial
design enables the experimenter to study the independent effect as well as interactive effect of two
or more independent variables

Terms related to factorial Design

Factors : The term factor is broadly used to include the independent variable that is
manipulated by the investigator in the experiment or that is manipulated through selection.
In the research some time it is possible to manipulate the independent variable directly, for
example in a study researcher wants to study the effect of different drugs on the recovery
of the patient. The researcher may select three dosages 2 mg, 4 mg. and 6 mg. and
administer the drug to the subjects. Further researcher may find that age is another
important variable that may influence the rate of recovery from the diseases. The second
independent variable that is age cannot be directly manipulated by the researcher. The
manipulation of the variable ‘age’ is achieved through selection of the sample. The
researcher then may divide the subjects into three age groups.

Main Effect: This is the simplest effect of a factor on a dependent variable. It is the effect of
the factor alone averaged across the level of other factors. 18 Research Design According to
Mcburney & White (2007) main effect in a factorial experiment, the effect of one
independent variable, averaged over all levels of another independent variable.

Interaction: The conclusion based on the main effects of two independent variables may be
at times misleading, unless we take into consideration the interaction effect of the two
variables also. According to Mcburney & White (2007) Interaction means when the effect of
one independent variable depends on the level of another independent variable. An
interaction is the variation among the difference between mean for different levels of one
factor over different levels of the other factor.

For example a cholestrol reduction clinic has two diets and one exercise regime. It was
found that exercise alone was effective and diet alone was effective in reducing cholestrol
levels (main effect of exercise and main effect of diet). Also for those patients who didn’t
exercise, the two diets worked equally well (the main effect of diet); those who follow diet A
and exercised got the benefits of both (main effect of diet A & main effect of exercise).
However it was found that those patients who followed diet B and exercised got the benefit
of both plus a bonus, an interaction effect (main effect of diet B, main effect of exercise plus
an interaction effect of diet and exercise).

Types of interaction

1. Antagonistic interaction: This occurs when the main effect (the individual effect) of each
independent variable is not significant, meaning they don't have a noticeable impact on the
outcome by themselves. However, when these variables are combined, their interaction
becomes significant, and they tend to cancel out or reverse each other's effects.
Example: Let's say we're studying the effects of studying time and sleep on exam
performance. Individually, neither studying time nor sleep seems to have a significant
impact on exam scores. However, when we consider the interaction between studying time
and sleep, we find that students who study a lot and sleep less tend to perform worse than
students who study less but get more sleep. The combined effect of these two variables is
antagonistic because the positive effect of studying is counteracted by the negative effect of
sleep deprivation.

2. Synergistic interaction: In this type of interaction, a higher level of one independent

variable enhances the effect of another independent variable. In other words, the combined
effect of the two variables is greater than the sum of their individual effects.
Example: Let's consider the effects of exercise and diet on weight loss. Individually, exercise
and diet may have positive effects on weight loss, but when combined, their effects are
even more significant. People who exercise regularly and follow a healthy diet tend to lose
more weight than those who only focus on one of these factors. The interaction between
exercise and diet in this case is synergistic because they work together to produce a
stronger effect on weight loss.

3. Ceiling effect interaction: This occurs when the higher level of one independent variable
reduces the differential effect of another variable. In other words, one variable has a smaller
impact when paired with a higher level of another variable.
Example: Let's imagine a study on the effects of motivation and rewards on task
performance. Initially, increasing motivation leads to a significant improvement in
performance. However, when high levels of rewards are introduced, the impact of
motivation on performance diminishes. In this case, the ceiling effect interaction suggests
that the presence of high rewards reduces the additional benefit gained from increased
motivation. The effect of motivation reaches a ceiling, and further rewards do not
contribute as much to performance improvement.
These examples help illustrate how different types of interactions can occur between
independent variables, affecting the overall outcome or dependent variable in a study.
Type of factorial design

1. Within-Subject Factorial Design: In a within-subject factorial design, participants are

exposed to all combinations of the independent variables. This means that each participant
experiences every level of every independent variable. This design is also referred to as a
repeated measures or within-subjects design.
Example: A study examines the effects of two factors, lighting condition (variable A) and
background noise level (variable B), on participants' task performance (dependent variable).
Each participant is tested under all combinations of lighting condition and background noise
level.
Table for Within-Subject Factorial Design:

Participant Lighting Condition Background Noise Level Task Performance

P1 Low Low 80

P1 Low High 70

P1 High Low 85

P1 High High 75
 Participant Lighting Condition Background Noise Level Task Performance

P2 Low Low 90

P2 Low High 65

P2 High Low 88

P2 High High 72

In this design, each participant serves as their own control, and the focus is on comparing
performance across different combinations of lighting condition and background noise level.

2. Between-Subject Factorial Design: In a between-subject factorial design, different groups

of participants are assigned to each combination of the independent variables. Each
participant is only exposed to one level of each independent variable. This design is also
known as an independent groups or between-subjects design.
Example: A study examines the effects of two factors, dosage (variable A) and treatment
type (variable B), on pain reduction (dependent variable). Participants are randomly
assigned to one combination of dosage and treatment type.
Table for Between-Subject Factorial Design:

Participant Dosage Treatment Type Pain Reduction

P1 Low Drug A 5
 Participant Dosage Treatment Type Pain Reduction

P2 High Drug A 8

P3 Low Drug B 7

P4 High Drug B 9

In this design, different groups of participants are exposed to different combinations of

dosage and treatment type, and their pain reduction levels are measured.
3. Mixed Factorial Design: A mixed factorial design combines elements of both within-
subject and between-subject designs. It includes at least one independent variable that is
manipulated within-subjects and at least one independent variable that is manipulated
between-subjects.
Example: A study examines the effects of two factors, training method (within-subject) and
gender (between-subject), on reaction time (dependent variable). Each participant
undergoes both training methods, but the groups are divided by gender.
Table for Mixed Factorial Design:

Participant Training Method Gender Reaction Time

P1 Method A Male 300 ms

P1 Method B Male 275 ms

 Participant Training Method Gender Reaction Time

P2 Method A Female 350 ms

P2 Method B Female 325 ms

In this design, each participant experiences both training methods, but the groups are
separated by gender.
These examples and tables help illustrate the different types of factorial designs, how they
are structured, and how data can be presented for analysis.

Factorial designs: Advantages and Disadvantages

Pros of Factorial Design Research:

1. Efficient use of resources: Factorial designs allow researchers to study multiple
factors simultaneously, maximizing the efficiency of data collection. By manipulating
and measuring multiple variables in a single study, researchers can gather more
comprehensive and nuanced information.
2. Examination of interactions: Factorial designs enable the investigation of
interactions between independent variables. Interactions help researchers
understand how the effects of one variable may be influenced by another, providing
deeper insights into the complexity of real-world phenomena.
3. Generalizability: By manipulating multiple factors, factorial designs can capture a
broader range of conditions, making the findings more generalizable to real-world
settings. This enhances the external validity of the research.
4. Statistical efficiency: Factorial designs allow for the estimation of main effects and
interactions simultaneously, providing more statistically efficient analyses.
Researchers can obtain more precise estimates and draw stronger conclusions from
their findings.
Cons of Factorial Design Research:
 1. Increased complexity: Factorial designs can be more complex to plan, execute, and
analyse compared to single-factor designs. With multiple independent variables and
their levels, there is a greater need for careful experimental design and data
interpretation.
2. Sample size requirements: The inclusion of multiple factors and their levels may
increase the required sample size, leading to higher costs and logistical challenges.
Large sample sizes are often necessary to achieve sufficient statistical power for
detecting interactions.
3. Potential confounding: In factorial designs, there is a risk of confounding variables,
especially when the factors are not fully orthogonal (i.e., independent of each other).
Confounding variables can complicate the interpretation of results and limit causal
inferences.
4. Limited focus on individual factors: While factorial designs allow for the
examination of interactions, they may provide less detailed information about the
individual effects of each factor. Researchers may need additional studies or
analyses to gain a deeper understanding of the isolated effects of each factor.
It is important for researchers to carefully consider these pros and cons when selecting and
designing a factorial study, ensuring that the benefits outweigh the potential challenges and
limitations.
Advantages of Quasi-Experimental Design:
 Real-World Context: Quasi-experimental designs allow researchers to study
phenomena in real-world settings where random assignment may not be feasible or
ethical. This provides a greater degree of external validity, as the findings can be
more easily generalized to similar populations and settings.
 Practicality: Quasi-experimental designs often require fewer resources, time, and
logistical constraints compared to true experimental designs. Researchers can
leverage existing groups or naturally occurring events, making it a more practical
approach in many research situations.
 Increased Generalizability: Quasi-experimental designs allow for the inclusion of
diverse participant groups, enhancing the generalizability of the findings beyond a
specific sample.
 Ethical Considerations: In some cases, random assignment may not be possible due
to ethical concerns. Quasi-experimental designs provide an alternative approach that
allows researchers to investigate causal relationships while respecting ethical
boundaries.

Disadvantages of Quasi-Experimental Design:

 Threats to Internal Validity: Quasi-experimental designs are susceptible to various
threats to internal validity, such as selection bias, history effects, maturation, and
regression to the mean. Without random assignment, it becomes challenging to
attribute observed differences solely to the independent variable.
 Limited Control: Researchers have less control over the manipulation of the
independent variable in quasi-experimental designs. This lack of control increases
the risk of confounding variables influencing the results, potentially leading to biased
conclusions.
 Pre-existing Differences: Quasi-experimental designs often involve pre-existing
groups, which may already differ on certain variables that can impact the outcome.
This can make it difficult to determine whether the observed effects are solely due
to the independent variable or the pre-existing differences.
 Inability to Establish Causality: While quasi-experimental designs allow researchers
to investigate relationships between variables, they fall short of establishing a
definitive cause-and-effect relationship due to the absence of random assignment.
Other factors beyond the independent variable may contribute to the observed
outcomes.
 Limited Internal Validity: Quasi-experimental designs may lack the same level of
internal validity as true experimental designs. The inability to control for all potential
confounding variables and threats to internal validity makes it challenging to draw
strong causal inferences.