Science, Tech, Math › Math What Is a Double Blind Experiment? Share Flipboard Email Print Yuri_Arcurs/Getty Images Math Statistics Applications Of Statistics Statistics Tutorials Formulas Probability & Games Descriptive Statistics Inferential Statistics Math Tutorials Geometry Arithmetic Pre Algebra & Algebra Exponential Decay Functions Worksheets By Grade Resources View More By Courtney Taylor Professor of Mathematics Ph.D., Mathematics, Purdue University M.S., Mathematics, Purdue University B.A., Mathematics, Physics, and Chemistry, Anderson University Courtney K. Taylor, Ph.D., is a professor of mathematics at Anderson University and the author of "An Introduction to Abstract Algebra." our editorial process Courtney Taylor Updated February 04, 2019 In many experiments, there are two groups: a control group and an experimental group. The members of the experimental group receive the particular treatment being studied, and the members of the control group do not receive the treatment. Members of these two groups are then compared to determine what effects can be observed from the experimental treatment. Even if you do observe some difference in the experimental group, one question you may have is, “How do we know that what we observed is due to the treatment?” When you ask this question, you are really considering the possibility of lurking variables. These variables influence the response variable but do so in a way that is difficult to detect. Experiments involving human subjects are especially prone to lurking variables. Careful experimental design will limit the effects of lurking variables. One particularly important topic in the design of experiments is called a double-blind experiment. Placebos Humans are marvelously complicated, which makes them difficult to work with as subjects for an experiment. For instance, when you give a subject an experimental medication and they exhibit signs of improvement, what is the reason? It could be the medicine, but there could also be some psychological effects. When someone thinks they are being given something that will make them better, sometimes they will get better. This is known as the placebo effect. To mitigate any psychological effects of the subjects, sometimes a placebo is given to the control group. A placebo is designed to be as close to the means of administration of the experimental treatment as possible. But the placebo is not the treatment. For example, in the testing of a new pharmaceutical product, a placebo could be a capsule that contains a substance that has no medicinal value. By use of such a placebo, subjects in the experiment would not know whether they were given medication or not. Everyone, in either group, would be as likely to have psychological effects of receiving something that they thought was medicine. Double Blind While the use of a placebo is important, it only addresses some of the potential lurking variables. Another source of lurking variables comes from the person who administers the treatment. The knowledge of whether a capsule is an experimental drug or actually a placebo can affect a person’s behavior. Even the best doctor or nurse may behave differently toward an individual in a control group versus someone in an experimental group. One way to guard against this possibility is to make sure that the person administering the treatment does not know whether it is the experimental treatment or the placebo. An experiment of this type is said to be double blind. It is called this because two parties are kept in the dark about the experiment. Both the subject and the person administering the treatment do not know whether the subject in the experimental or control group. This double layer will minimize the effects of some lurking variables. Clarifications It is important to point out a few things. Subjects are randomly assigned to the treatment or control group, have no knowledge of what group they are in and the people administering the treatments have no knowledge of which group their subjects are in. Despite this, there must be some way of knowing which subject is in which group. Many times this is achieved by having one member of a research team organize the experiment and know who is in which group. This person will not interact directly with the subjects, so will not influence their behavior.