The Characteristics of an Experiment


What is an Experiment?

An experiment is a study in which a researcher assigns subjects to a specific group and then applies different treatments to each group. The purpose of an experiment is to understand a cause-and-effect relationship between a treatment and an outcome.

Terms to know

Response variable – a variable that we are interested in studying in an experiment.

Explanatory variable – a variable we use to explain any changes in the response variable.

Blind experiment – an experiment in which the subjects don’t know which treatment they are receiving. 

Double blind experiment – an experiment in which both the researchers and the subjects don’t know which treatment the subjects are receiving.

Treatment group – the group of subjects who receives a specific treatment that we’re interested in seeing the effects of.

Control group – the group of subjects who does not receive a treatment. We use this group to compare the difference in outcomes between the subjects who received the treatment and those who did not. 

Experimental unit – the entity that we assign to a treatment. 

Confounding variable – a variable we do not account for in an experiment that could make the results of an experiment unreliable or useless.

Example 1:

Researchers want to know if eating spinach on a regular basis impacts blood pressure. Researchers split up 200 people into two groups. One group is assigned to a spinach-diet and one group is assigned to their normal diet. After one month, researchers record the beginning and ending blood pressure of each person. 

Response variable – the change in blood pressure.

Explanatory variable – the diet.

Blind experiment – No. The subjects know which group they are in.

Double blind experiment – No. Both the researchers and subjects know which group the subjects are in.

Treatment group – the group of subjects who is assigned to the spinach-diet.

Control group – the group of subjects who is assigned to their normal diet.

Experimental unit – the individual people.

Confounding variables – Variables that we did not account for that may impact the results of the experiment could include age and time spent exercising by the individuals. Both of these variables could impact blood pressure and we are not accounting for them in the experiment.

Example 2:

Researchers want to know if a new studying technique helps students do better on a math test. Researchers give 100 students a math test. Then, the researchers split up 100 students into two groups. One is assigned to use the new studying technique for two weeks and the other is assigned to use their normal studying technique for two weeks. After two weeks of studying, the researchers give another math test to all the students and record their beginning and ending scores.

Response variable – the change in test score.

Explanatory variable – the studying technique.

Blind experiment – No. The subjects know which group they are in.

Double blind experiment – No. Both the researchers and subjects know which group the subjects are in.

Treatment group – the group of subjects who is assigned use the new studying technique.

Control group – the group of subjects who is assigned to use their normal studying technique.

Experimental unit – the individual students.

Confounding variable – A variable that we did not account for that may impact the results of the experiment could be time spent studying. This could impact the final test results and we are not accounting for them in the experiment.

Example 3:

Researchers want to know if a new drug causes weight loss. A third party recruits 100 people and randomly assigns 50 to take the new drug once per day and 50 to take a placebo (a harmless sugar pill that looks like the real drug but actually does nothing) once per day. The subjects do not know if they’re taking the drug or the placebo. After two months, researchers record the beginning and ending weight of each person.

Response variable – the change in weight.

Explanatory variable – the new drug.

Double blind experiment – Yes. Both the researchers and subjects are unaware of which group the subjects are in.

Treatment group – the group of subjects who takes the new drug.

Control group – the group of subjects who takes the placebo.

Experimental unit – the individual people.

Confounding variables – Variables that we did not account for that may impact the results of the experiment could include age and time spent exercising by the individuals. Both of these variables could impact blood pressure and we are not accounting for them in the experiment.

Leave a Reply

Your email address will not be published. Required fields are marked *