Experiment Design

What is the Experiment Design?

Experiment design also known as the design of the experiment is a division of applied statistics that takes care of planning, conducting, analysing and explaining controlled tests to estimate the factors that control the value of the parameter of a group of parameters. The design of experiment is an influential collections and analysis tool that can be used in multiple experimental situations.

It enables various input factors to be manipulated, identifying their effect on the desired output (response). By managing various inputs at the same time, experiment design can determine essential interactions that may be missed while experimenting with a single factor at a time. All possible combinations can be examined (full factorial) or only part of the possible combinations (fractional factorial).

[Image will be Uploaded Soon]

Experimental Design Definition

In Statistics, the experimental design is defined as the design of a research experiment in which we determine whether a variation is present or not, and it should be analysed  under the complete control of the researcher. The experiment design is primarily used for controlled experiments. These experiments diminish the effects of the variable to enhance the reliability of outcomes. In the experiment design, the process of an experimental unit may involve a group of people, plants, animals, etc.

What is the Experimental Research Design?

Experimental research design is research performed with a scientific approach with the help of the two sets of variables. The first set is determined as a constant, which you use to calculate the differences of the second set. It is necessary to determine the facts if you don't have sufficient data to support your decision. Experimental research collects the necessary data to enable you to make optimum decisions. Any research performed under scientifically acceptable conditions makes use of experimental methods. The success of experimental studies relies on the researchers confirming the variations of a variable which is simply based on the directions of the constant variable. The research should determine a significant cause and effect.

Experiment Research Design Examples

Some of the experimental research designs are as follows:

There are different experimental research design examples. It relies on the types of experimental research design that has been considered. The important experiment research design example is laboratory experiments which may differ in nature depending on the topic of the research.

Some more experimental research designs examples are as follows:

  • Evaluation of exams after the end of the semester

  • Evaluation of employees skill in organization

  • Evaluation of teaching methods of different teachers by the schools.

Types of Experiment Research Design

Here are some of the different types of experimental research designs.

  • Pre-experimental Research Design

  • True-experimental Research Design

  • Quasi-Experimental Research Design

1. Pre-experimental Research Design

The pre-experimental research design is the simplest method of research design in statistics.In this method, a single group or multiple groups are kept under observations, after some factors are identified for the cause and effect. This method is mostly  performed in order to understand whether further investigations are required for the targeted group. Due to this, this method is considered to be cost-effective. The pre-experimental research design method is divided into three types, namely,

  • Static-group Comparison

  • One- group Pretest-posttest Experimental Research Design

  • One-shot Case Study Experimental Research Design

2. True-experimental Research Design

True-experimental research design is the most appropriate form of experimental research design as it depends on the statistical hypothesis to prove or disprove the hypothesis. This method is primarily used in Physical Science. It is the only method that determines the cause and effect relationship within the groups. The important factors which are required  to be satisfied in this particular method are:

  • Random variable

  • Variable can be easily managed by the researcher

  • Control Groups (A group of participants are informed about  the experimental group, but the experimental rules do not apply to them)

  • Experimental Group (Research participants where experimental rules are used)

3. Quasi-Experimental Design

The word "quasi" means partial, half, or pseudo. A quasi-experimental design is the same as the true experimental design, but there are few differences between them.

In true experiment design, the participants of the group are appointed randomly. So, every unit has an equal opportunity to participate in the experimental group. Whereas, in a quasi-experimental design, the participants of the groups are not appointed randomly So, the researcher cannot make a cause or effect conclusion that cannot be made by the researcher. Thus, it is not possible to appoint the participants into the group.

This method is very commonly used in educational research, where administrators are forced to permit the random selection of students for experimental samples. Some of the examples of quasi-experimental research design are the time series, no equivalent control group design, and the counterbalanced design.

Solved Examples

1. A Latin Square Design was used to Compare the Four Diets  ( P, Q, R, S ) with 4 Cows in 4 Different Periods.


                      1                       2                     3                    4






















a. Write Out the ANOVAs Test and Indicate F Tests.

b. Assume the Diet was as Follows, What Orthogonal Contrast will you use to Test Main Effects and Interactions.

P – First cut alfalfa, cut in early bud stage

Q - First cut alfalfa, cut in early bloom stage

R - First cut alfalfa, cut in early bud stage

S - First cut alfalfa, cut in early bloom stage


Latin square columns= cows, rows = period















b. Treatment are arranged in a 2 * 2 factorial

  Contrasts :

  • Main effect of cut: P + Q + R +S

  • Main effect of stage: P + Q + R +S

  • Interaction: P + Q + R +S

2. The Effect of Each of Three Preparations on the Glycogen Content of the Liver was Studied as Follows:

Each of these preparations ( control compound 217, compound 217 + sugar,) was used on two different rabbits ( and for a total 6 rabbits) and four estimations of liver glycogen were made per rabbit.

The objective of this experiment is to determine if there is a significant difference among the three treatments (preparation) and to estimate the two variance components( reading with rabbits or sampling errors, and rabbits within a treatment or experimental error

a. Examine the experimental units and sampling units,.How many replicates of each treatment are there? Are the treatments fixed or random?

b. Complete the analysis of the variance table and indicate appropriate F-tests.


Random sampling within rabbits

Experimental units - rabbits

Sampling units - glycogen tests

Glycogen test- 2

Treatments are fixed





Rabbit within Trt


Glycogen within rabbit




Quiz Time

1. What is the Meaning of the Replication of Experiment?

  1. Only a few treatments are applied

  2. Comprehensive performance of the experiment. That ts each treatment possibility is applied.

  3. None of the above

2. In a Completely Randomized Experiment, Each of the Runs is Made in Random Order.

  1. True

  2. False

3. Which of these is Used to Start a New Process?

  1. Design of experiment

  2. Control charts

  3. Histograms

  4. Acceptance of sampling

4. The Experiment Design is Used to Examine the Variables Which are _________ Affecting the State of the Process?

  1. The most

  2. The least

  3. Not

  4. Not changing or

Fun Facts

  • Francis Bacon was the first to declare a true scientific method of experimental design  but he didn't perform in vaccum.

  • The statistical principles basically the design of the experiment was largely introduced by R.A. Fisher during his pioneering work at the Rothamsted Experimental Station in the 1920s and 1930s.

FAQ (Frequently Asked Questions)

1. What are the Advantages and Disadvantages of Experimental Research Designs?

Advantages of experimental research designs are as follows:

  1. Researchers have a strong set back to get desired results.

  2. The results are expressive.

  3. Researchers can apply their findings to similar ideas or circumstances after analyzing the results.

  4. The subject or industry does not affect the potency of experimental research. Any industry can use it for research purposes.

  5. The cause and effect of the hypothesis can be identified easily. Researchers can further examine the relationship to identify more detailed ideas.

Disadvantages of experimental research designs are as follows:

  1. It is costly . 

  2. The risk factor is high in experimental research design and may include ethical implications that cannot be ignored. This is ordinary in medical research, where failed trials may bring about a patient's death or a worsening health condition.

  3. Experimental research results are not expressive.

  4. Response bias can also be delivered by the topic of the conversation.

  5. Human responses are hard to measure in  experimental design.

What are the three Principles of Experimental designs?

The three principles of experimental designs are :

  1. Randomization

  2. Replication

  3. Blocking

Randomization - It is the random requesting of experiments to ensure each level of a factor has an equal opportunity of being affected by noise factors such as the  temperature of power fluctuation.

Replication - It is the process of repeating all or portions of experiments that operate in a random sequence to permit a more accurate estimation of experimental errors along with main and interaction effects.

Blocking- it is the process of arranging the same experiment runs into groups or blocks to allocate the effect of change in blocking factors such as batch, machine, time of day etc. across the experiment and avoid confusion whether the variation in output is due to change in factor level or blocks.