Suggested languages for you:

Americas

Europe

|
|

# Conducting an Experiment

Have you ever tried something different or new just to see what would happen as a result? For example, did you ever change some ingredients in a recipe to see if you could make a fluffier cake? Or did you ever try a new hand cream to see if it helped to soothe your dry skin? Most of us will answer yes to these questions. In a way, these are experiments we do in our daily lives. Whenever we try something to see what will happen or make predictions to see whether the outcomes will match our predictions, we are simply conducting an experiment.

In this article, you will learn what an experiment is, its main components, and the characteristics of a well-designed experiment. Read on if you are feeling experimental!

## What is an Experiment?

An experiment is a type of study carried out to investigate the cause-and-effect relationship between two or more variables.

The purpose of conducting an experiment is to find evidence to decide whether that cause-and-effect relationship exists or not, but how exactly do you do this? See the following section to better understand what is involved in conducting an experiment.

## Conducting an Experiment Meaning

To be able to understand the meaning of conducting an experiment, you need to familiarize yourself with the terminology used in this type of study.

### Roles in an Experiment

First, let's look at the different roles involved in conducting an experiment in the table below.

 Role Definition Example Researcher(s) People involved in the planning and execution of the experiment. A researcher or group of researchers working together in an experiment. Experimental Units These are the units that are assigned a treatment, as part of an experiment.If the experimental units are people, they are called subjects or participants. These can be people, animals, plants, or objects. Evaluators If required, some experiments might involve evaluators, who are responsible for evaluating the results of the experiment, providing an unbiased and reliable evaluation of the outcomes obtained. A senior/more experienced group of people within a research group.
Table 1. Roles in experiments, conducting an experiment.

### Types of Variables in an Experiment

Other important terms that you need to be aware of when conducting experiments are related to the different types of variables involved in the process, these are described in the following table.

 Type of Variable Definition Example Independent / Explanatory Variable / Factor The independent or explanatory variable (also known as a factor) is the variable that you manipulate in an experiment. It is the variable that you can control.It is independent because its value is not dependent on, or affected by, any other variable in the experiment. In an experiment to investigate the effect of a new energy drink on race performance. The explanatory variable would be the amount of energy drink consumed by the participants. Treatments / Experimental Conditions The treatments (also known as experimental conditions) are the different levels of a factor, or combinations of levels, in the case when two or more factors are considered, that an experimental unit is assigned. In the same experiment, some possible treatments could be:Treatment $$1$$: Consume no energy drink at all,Treatment $$2$$: Consume half an energy drink,Treatment $$3$$: Consume $$1$$ energy drink. Dependent / Response Variable The dependent variable is the variable that you measure in an experiment. The dependent variable is sometimes referred to as the response variable, as it responds to changes in the independent variable. The dependent variable in the same experiment would be the race performance of the participants (i.e. time taken to run $$100\ m$$). Confounding Variable A confounding variable is a variable that is related to the explanatory variable, and it also affects the response variable, making it difficult to determine which one of the two is really causing the response. Some possible confounding variables in this experiment could be the participants' age, diet, level of exercise, etc., which are related to the amount of energy drinks that a person consumes, and also to their performance in a race.

Table 2. Types of variables in experiments, conducting an experiment.

To help you visualize the concepts more clearly, Figure 1 illustrates the main difference between the explanatory variable and the response variable in an experiment.

Fig. 1. Explanatory vs response variable

Now, that you have a better idea of the terminology involved in conducting an experiment, you can define the meaning of conducting an experiment as follows:

Conducting an experiment is the process where researchers manipulate an explanatory variable to define treatments, which are randomly assigned to experimental units or subjects, to then compare the responses of the different groups to the treatments received.

In simple terms, an experiment is a process of testing a hypothesis with the aim of accepting or rejecting it. Researchers often develop hypotheses when conducting scientific experiments to guide experimental design.

A hypothesis is an assumption of what will happen as a result of an experiment.

## Characteristics of a Well-Designed and Conducted Experiment

A well-designed and conducted experiment should have the following characteristics:

 Characteristic Description Comparison You will need to have at least two groups to be able to compare their response. Those groups should comprise:a control group (that receives no treatment or receives a placebo) to be used as a baseline, and one or more experimental groups (that receive a treatment). This is to be able to compare the response of those groups receiving the treatments with the one of the control group, thus seeing more clearly the effect of the treatments on the experimental units or subjects. Random Assignment Experimental units or subjects are assigned to treatments at random. Replication A well-designed and conducted experiment should be able to be replicated in different groups. Control Controlling events or conditions surrounding an experiment is very important for a well-designed experiment. An experiment with control is set to ensure that the response is the result of the independent variables in the study and not the result of any other source of variability.Control involves making sure that all participants in an experiment are treated equally, and that they are exposed to the same conditions. The only difference between groups in an experiment should be the treatment that they are receiving.

Table 3. Experiment characteristics, conducting an experiment.

A placebo is used in well-designed experiments to eliminate the placebo effect, which refers to the response that individuals can have to treatment just because they are expected to have one, even when they are not given any treatment at all. The placebo is a treatment that is meant to cause no effect on the subject. Its purpose is to make the conditions the same for all groups involved in an experiment and to be able to differentiate the response from just receiving treatment from the one related to the effectiveness of the treatment itself.

## Conducting an Experiment Steps

The steps involved in conducting an experiment are as follows:

1. State the question that the experiment intends to answer.

This can be a question or a hypothesis. As mentioned before, a hypothesis is an assumption of what will happen as a result of an experiment. The words "If" and "Then" are often used in describing or writing hypotheses. For example, "If I do not expose my plants to sunlight, then they will die". The "If" and "Then" statements suggest the independent and dependent variables.

2. Specify the response variable.

What is the dependent variable that you will measure? and how will you measure the response?

3. Specify the factor levels/treatments.

Depending on the number of factors involved in the experiment, what levels of those factors will be used as treatments?

4. Define the experimental units.

Who or what will be receiving the treatments?

5. Make sure that the characteristics of a well-designed and constructed experiment are followed (comparison, random assignment, replication, and control).

6. Draw a diagram.

7. Data collection.

Carry out the experiment and collect the data.

8. Analysis of results.

Analyze and compare the data obtained from the response of the different groups to their corresponding treatments, so that the initial question/hypothesis can be answered. This step might be carried out by the evaluators if required.

## Example of Conducting an Experiment

Now that you are familiar with the terminology and the steps for conducting a well-designed experiment, let's look at an example.

Imagine the following hypothetical scenario: A group of researchers is carrying out an experiment to investigate if a new drug, called $$Drug\ A$$, can provide faster pain relief for the treatment of headaches than other available drugs in the market. Follow the steps to conduct a well-designed experiment.

Solution:

To solve this, you can go through steps $$1 - 6$$ from the previous section. Steps $$7$$ and $$8$$ can be omitted as they require carrying out the experiment.

1. State the question that the experiment intends to answer.

In this example, the question is: Does $$Drug\ A$$ provide faster pain relief for the treatment of headaches than other available drugs in the market?

2. Specify the response variable.

The response variable will be the time that it takes the participants to get effective pain relief from their treatment. This variable will be measured in minutes.

3. Specify the factor levels/treatments.

You need to set up a placebo treatment for the control group to be used as a baseline. In this case, you can use a "sugar pill" that looks as similar as possible to the other drugs being used in the experiment. This is to eliminate the placebo effect so that all participants feel part of the experiment.

The explanatory variable or factor in this example will be the type of pain relief drug given to the participants.

The treatments will be as follows:

1. Group $$1$$ will get the placebo treatment of a "sugar pill".
2. Group $$2$$ will get a dose of a regular pain relief drug.
3. Group $$3$$ will get a dose of $$Drug\ A$$.

4. Define the experimental units.

The experimental units, in this case, subjects or participants, will be three groups of $$10$$ people to be assigned the $$3$$ treatments at random. The groups of people should be comprised of people of a similar range of ages, and gender, to avoid the effect of confounding variables on the response to the treatments.

5. Make sure that the characteristics of a well-designed and constructed experiment are followed (comparison, random assignment, replication, and control).

• Comparison - $$3$$ groups are receiving different treatments to be able to compare their response.
• Random Assignment - The treatments will be assigned to the different groups at random.
• Replication - $$10$$ participants in each group will receive treatment.
• Control - The same conditions will be provided to all $$3$$ groups so that their environment doesn't affect their response to the treatment.

6. Draw a diagram.

Figure 2 shows the diagram illustrating the different components of the pain relief experiment, including participants, groups, treatments, and response variable.

Fig. 2. Diagram of a pain relief experiment

One more example.

Going back to the scenario from the introduction of this article, let's design an experiment to investigate how to make a fluffier cake. To do this, you can try adding one more egg to the original recipe.

Solution:

1. State the question that the experiment intends to answer.

Does adding one more egg to a cake recipe result in a fluffier cake?

2. Specify the response variable.

Consistency of the cake: is it fluffier?

3. Specify the factor levels/treatments.

In this case, the explanatory variable or factor that you will manipulate will be the number of eggs in the recipe of cake.

The treatments will be as follows:

1. Group $$1$$ will be the original recipe, to be used as a baseline.
2. Group $$2$$ will the original recipe but add an additional egg.

4. Define the experimental units.

The experimental unit in this case will be the cake batter.

5. Make sure that characteristics of a well-designed and constructed experiment are followed (comparison, random assignment, replication, and control).

• Comparison - $$2$$ groups of $$3$$ cakes each will be baked using $$2$$ different recipes to be able to compare their results.
• Random Assignment - The same ingredients will be used for all cakes and distributed among each group, ensuring that they are all checked for the quality and quantities required.
• Replication - $$3$$ cakes will be baked in each group.
• Control - Experimental conditions will be the same for all cakes being baked including the type of oven used, cooking time, and temperature so that no other confounding variable affects the results.

6. Draw a diagram.

Figure 3 shows the diagram illustrating the different components of the cake recipe experiment, including experimental units, groups, treatments, and response variable.

Fig. 3. Diagram of a cake recipe experiment

## Importance of Conducting an Experiment

Some of the reasons why conducting experiments are of vital importance are as follows:

• Allows finding evidence about cause-effect relationships between two or more variables.
• Explains change.
• Demonstrates or proves a hypothesis to be able to accept it or reject it.
• Determines the effects of something previously unconfirmed.

Experiments are a powerful method of data collection that facilitates important research projects that benefit everything around us, finding the reasons why something might be happening, better ways to do things, and making new discoveries.

Experiments are carried out in many fields, including medicine, science, technology, and engineering, among many other areas. They are part of our daily lives, and they play an important role in learning about our environment. So, let's carry on experimenting!

## Conducting an Experiment - Key takeaways

• An experiment is a type of study carried out to investigate the cause-and-effect relationship between two or more variables.
• The different roles involved in conducting an experiment are researchers, experimental units, and evaluators.
• The different types of variables involved in an experiment are explanatory variable, treatments or experimental conditions, response variable, and confounding variable.
• Conducting an experiment is the process where researchers manipulate an explanatory variable to define treatments, which are randomly assigned to experimental units or subjects, to then compare the responses of the different groups to the treatments received.
• The characteristics of a well-designed and conducted experiment are comparison, random assignment, replication, and control.

Conducting an experiment is the process where researchers manipulate an explanatory variable to define treatments, which are randomly assigned to experimental units or subjects, to then compare the responses of the different groups to the treatments received.

The steps involved in conducting an experiment are:

1. State the question that the experiment intends to answer.

2. Specify the response variable.

3. Specify the factor levels/treatments.

4. Define the experimental units.

5. Make sure that the characteristics of a well-designed and constructed experiment are followed (comparison, random assignment, replication, and control).

6. Draw a diagram.

7. Data collection.

8. Analysis of results.

The purpose of conducting an experiment is to find evidence to decide whether that cause-and-effect relationship exists or not.

Make sure that characteristics of a well-designed and constructed experiment are followed (comparison, random assignment, replication, and control).

Do:

• State a clear question of what the experiment intends to answer.
• Define the explanatory and response variables.
• Specify the treatments.
• Eliminate the placebo effect.
• Make sure that characteristics of a well-designed and constructed experiment are followed (comparison, random assignment, replication, and control).

Don't:

• Let participants choose their own treatment.
• Allow confounding variables to affect the response in an experiment.
• Set up different conditions for each group.
• Set up groups with different characteristics and size.
• Use only one experimental group.
• Have no control group with a placebo treatment or no treatment at all.

## Final Conducting an Experiment Quiz

Question

What is bias in experiment?

Bias in experiments refers to a known or unknown influence in the experimental process, data or results.

Show question

Question

What is the placebo effect?

A placebo is a medicine or procedure that has no active substance and no real effect. The placebo effect involves receiving a treatment that causes improvement even when its fake.

Show question

Question

What is blinding?

Blinding means to keep information from someone about the type of treatment they are getting.

Show question

Question

What is double blinding?

Double blinding is when both the participant and the person administering the treatment does not know if its a placebo or not.

Show question

Question

What is the reason for blinding?

Blinding is used to ensure that the patient or participant is not aware that they are being given a placebo treatment.

Show question

Question

List some sources of bias in experiments.

1. The method of data collection and the source of the data can lead to bias in experiments. To learn about the methods of data collection, see our article on Methods of Data Collection.
2. Not considering all possible outcomes can lead to bias.
3. Unknown changes in the experimental environment can lead to bias.
4. Adjusting the experiment results to fit an hypothesis is bias.
5. False behavior and response from the participants can lead to bias.

Show question

Question

List some types of bias.

1. Participant or Selection Bias.
2. Publication Bias.
3. Confirmation Bias
4. Observation Bias
5. Confounding Bias.
6. Design Bias.

Show question

Question

List some types of bias.

1. Participant or Selection Bias.
2. Publication Bias.
3. Confirmation Bias
4. Observation Bias
5. Confounding Bias.
6. Design Bias.

Show question

Question

List some ways of avoiding bias in experiments.

1. Ensure that the participants in your experiment represents represent all categories that are likely to benefit from the experiment.
2. Ensure that no important findings from your experiments are left out.
3. Consider all possible outcomes while conducting your experiment.
4. Make sure your methods and procedures are clean and correct.
5. Seek the opinions of other scientists and allow them review you experiment. They maybe able to identify things you have missed.
6. Collect data from multiple sources.
7. Allow participants to review the conclusion of your experiment so they can confirm that the conclusion accurately represents what they portrayed.
8. The hypothesis of an experiment should be hidden from the participants so they don't act in favor or maybe against it.

Show question

Question

List some advantages of eliminating bias in experiments.

1. The results and conclusion of the experiment will be reliable and dependable.
2. There will be better chances of the experiment helping as much people as it should.
3. Important information and findings will not be hidden or left out.
4. The conclusion of the experiment will not be influenced by any specific opinion.
5. The scientist will be open minded and consider all possibilities while conducting the experiment.
6. The data collected will be more accurate.
7. Detailed and complete articles and journals for the experiment will be published.

Show question

Question

Explain the participant bias.

Participant bias occurs when a certain group of people are selected to participate in an experiment or research. This group of people maybe of the same age, same gender or may have the same characteristics or behavior. The problem here is that only one category of the population is considered. The experiment will not cover the effect on the rest of the population.

Show question

Question

Explain Participant bias.

Participant bias occurs when a certain group of people are selected to participate in an experiment or research. This group of people maybe of the same age, same gender or may have the same characteristics or behavior. The problem here is that only one category of the population is considered. The experiment will not cover the effect on the rest of the population.

Show question

Question

What is publication bias?

Publication bias occurs when only the positive or interesting aspect of a scientific study is published.

Show question

Question

What is confirmation bias?

Confirmation bias is when you seek information that only helps you confirm your hypothesis without considering other possibilities.

Show question

Question

What is observation bias?

Observation bias is when the participants of an experiment exhibit false behavior because they know they are being watched.

Show question

Question

What is confounding bias?

Confounding bias is a type of bias that is as a result of an external factor affecting the relationship or association between a variable or subject that is being studied and its outcome.

Show question

Question

What is design bias?

Design bias is a bias that affects the outcome or conclusion of an experiment due to the procedures followed while conducting the experiment.

Show question

Question

What is an experiment?

An experiment, is a type of study carried out to investigate the cause-and-effect relationship between two or more variables.

Show question

Question

What are the experimental units in an experiment?

These are the units that are assigned a treatment, as part of an experiment. These can be people, animals, plants or objects.

Show question

Question

If the experimental units are people, they are called ________ or ________.

Subjects.

Show question

Question

What is the role of the people involved in the planning and execution of an experiment?

Researchers.

Show question

Question

True/False: The independent or explanatory variable (also known as a factor) is the variable that you manipulate in an experiment.

True.

Show question

Question

True/False: The dependent or response variable is the variable that you measure in an experiment.

True.

Show question

Question

A _________ variable is a variable that is related to the explanatory variable, and it also affects the response variable, making it difficult to determine which one of the two is really causing the response.

Confounding

Show question

Question

What is a hypothesis?

A hypothesis is an assumption of what will happen as a result of an experiment.

Show question

Question

A _______ is a treatment that is meant to cause no effect on the subject.

Placebo

Show question

Question

What is the placebo effect?

The placebo effect refers to the response that individuals can have to a treatment just because they are expected to have one, even when they are not given any treatment at all.

Show question

Question

True/False: Conducting an experiment is the process where researchers manipulate an explanatory variable to define treatments, which are randomly assigned to experimental units or subjects, to then compare the responses of the different groups to the treatments received.

True.

Show question

Question

The __________ are the different levels of a factor, or combinations of levels, in the case when two or more factors are considered, that an experimental unit is assigned.

Treatments

Show question

Question

The treatments in an experiment are also known as?

Experimental conditions.

Show question

Question

True/False: The characteristics of a well-designed and conducted experiment are: comparison, random assignment, replication and control.

True.

Show question

Question

What are experiment methods?

Experiment methods are the different methods that can be used to design and conduct an experiment.

Show question

Question

True/False: An experiment is a type of study carried out to investigate the cause-and-effect relationship between two or more variables.

True.

Show question

Question

What is random assignment?

Random assignment refers to assigning the treatments to the experimental units or subjects at random.

Show question

Question

What is the minimum amount of experimental groups that you are required to have in a well-designed experiment?

At least two.

Show question

Question

True/False: Each experimental group should contain a good variety of experimental units or subjects.

True.

Show question

Question

Which one of the following is a method for random assignment?

Draw numbers without replacement.

Show question

Question

What are the methods for random assignment?

Methods for random assignment include:

• Random number generator,
• Random number table,
• Drawing numbers without replacement.

Show question

Question

A ____________ experiment refers to an experiment where either the subjects, or the members of the research team measuring the response to the treatments, do not know what treatment the subjects are receiving.

Single-blind.

Show question

Question

A ___________ experiment refers to an experiment where both the subjects and the members of the research team measuring the response to the treatments, are unaware of what treatment the subjects are receiving.

Double-blind.

Show question

Question

True/False: A control group is a group of experimental units or subjects that receives no treatment or receives a placebo to be used as a baseline.

True.

Show question

Question

What is a placebo?

A placebo is a treatment that is meant to cause no effect on the subject.

Show question

Question

True/False: About $$20\%$$ of the subjects or participants in an experiment show a response to a placebo treatment due to the placebo effect.

True.

Show question

Question

What is blocking in randomized block design?

Blocking is an experiment method that involves dividing the subjects or participants in an experiment into blocks, based on a similar blocking variable. After this, each block of similar participants will be assigned their treatments at random.

Show question

Question

__________ is a good choice of method in matched pairs design, as there are only two members in each pair.

Flipping a coin.

Show question

Question

Within the randomized block design context, a block refers to a ____.

group, or collection.

Show question

Question

Within the randomized block design context, blocking is also known as ____.

stratifying.

Show question

Question

This is a variable that affects the outcomes of an experiment, but is not of particular interest for the experiment.

Nuisance variable.

Show question

Question

A brewery wants to test different amounts of yeast in their beverages, so they run an experiment each day across a week. Naturally, the temperature within each day varies. Which of the following is a nuisance variable?

The day's temperature.

Show question

Question

What is the difference between a randomized block design and a completely randomized design?

The difference is that in the randomized block design you first divide the population into groups before taking random samples.

Show question

60%

of the users don't pass the Conducting an Experiment quiz! Will you pass the quiz?

Start Quiz

## Study Plan

Be perfectly prepared on time with an individual plan.

## Quizzes

Test your knowledge with gamified quizzes.

## Flashcards

Create and find flashcards in record time.

## Notes

Create beautiful notes faster than ever before.

## Study Sets

Have all your study materials in one place.

## Documents

Upload unlimited documents and save them online.

## Study Analytics

Identify your study strength and weaknesses.

## Weekly Goals

Set individual study goals and earn points reaching them.

## Smart Reminders

Stop procrastinating with our study reminders.

## Rewards

Earn points, unlock badges and level up while studying.

## Magic Marker

Create flashcards in notes completely automatically.

## Smart Formatting

Create the most beautiful study materials using our templates.