Quasi-experimental research: what is it and how is it designed?

Quasi-experimental research is a type of research widely used in psychology.. Its most relevant characteristic is that the experimental groups are not selected randomly, but already formed groups are chosen (for example, a soccer team).

It is based on a descriptive methodology and some quantitative and qualitative elements, and is used to study different behaviors, social variables, etc. In this article we will know its characteristics and some differences with experimental research, as well as the advantages and disadvantages it presents.

• Related article: "The 15 types of research (and characteristics)"

What is quasi-experimental research?

Quasi-experimental research is used especially in the field of psychology, but also in the social sciences. This is a type of research halfway between experimental research and observational research. In reality, many authors do not consider it to be scientific, although it has notable advantages, as we will see in this article.

Unlike experimental research, in quasi-experimental research

the degree of control of extraneous variables (VVEE) is lower. On the other hand, the strange variables are those variables or factors that produce an effect on the variable that we are studying (variable dependent), but that we must control, since its effect is different from that produced by the independent variable(s) (which are the ones that interested in studying).

How is it investigated?

But how is it really investigated? Both in quasi-experimental research and in other types of research, whether in psychology or in other sciences, research It is mainly based on studying the effect of an independent variable (VI) (or more) on another variable, called the dependent variable (DV) (or more).

For example, we do research when we want to study the efficacy of a treatment (independent variable) in reducing anxiety (dependent variable).

Origin

quasi-experimental research originated in the educational field. It was born as a result of observing that some effects or phenomena could not be studied using the experimental method, and alternative designs had to be used. It was mostly about social phenomena or variables.

In recent years, the number of studies conducted through quasi-experimental research has been increasing more and more.

Characteristics

There are certain characteristics that differentiate quasi-experimental research from other types of research. They are the following.

1. no randomness

The basic characteristic of quasi-experimental research (and what differentiates it from strictly experimental research) is the non-randomness in the formation of the experimental groups. That is, the researcher selects already formed groups (for example, the students of a course or the workers of an office) to carry out his experiment.

In addition, this type of research is used when subjects cannot be randomly assigned to different experimental conditions Of the investigation.

Example

To illustrate, let's think of an example: let's imagine that we want to study the efficacy of three types of psychological therapy (for example psychodynamic, cognitive-behavioral and systemic) in reducing the level of anxiety in a group of people.

If we used an experimental design and not a quasi-experimental one, we would assign the subjects to the different experimental conditions (in this case, the three types of therapy) randomly, that is, using the random.

In quasi-experimental research, on the other hand, we could not do that. To solve this problem, frequently it is decided to include a control group in the experiment.

2. No control for secondary systematic variance

On the other hand, quasi-experimental research also a good choice when secondary systematic variance cannot be controlled; This originates when the internal validity of the experiment is threatened. Internal validity is what ensures that the independent variable is the cause of the dependent variable (that is, exerts an effect on it).

• You may be interested in: "The 10 Fundamental Types of Validity in Science"

Consequences

When a type of quasi-experimental research is used, and since the experimental groups have not been selected randomly, one thing happens: that we cannot guarantee that all subjects have similar characteristics. That is, there is less control of the variables. This makes the results less reliable (hence the name "quasi" experimental).

This means that this type of research is not used as much in laboratory contexts, but rather in natural contexts, in schools, etc. In other words, it is used above all in applied research.

Thus, quasi-experimental research has both positive and negative components. Let's see its advantages and disadvantages.

Advantages

The main advantage of quasi-experimental research is that Allows you to select accessible and already formed groups; In addition, it is often difficult to find groups that meet all the requirements to participate in an experiment (as would occur in an experimental design).

On the other hand, they are easy to apply and cheap designs. The preparation time they require and the resources to allocate are less than in an experimental design. In addition, it is a type of research that can be applied not only to study groups, but also individual cases.

Disadvantages

As negative characteristics or disadvantages in quasi-experimental research, we find its lower precision and lower validity, compared to experimental designs.

In addition, the lack of randomness in the formation of the groups poses a threat to the validity of the experiment and its precision or accuracy.

On the other hand, many times in this type of experiments the so-called placebo effect occurs, which consists of feeling or perceiving an improvement after believing that we have received a treatment (which in reality we have not received).

• You may be interested in: "What is the placebo effect and how does it work?"

types of designs

In quasi-experimental research, specifically in the field of psychology, two types of quasi-experimental designs are especially used:

1. cross-sectional designs

From these designs different groups are studied at a specific time point. For example, we can use them to measure the intelligence quotient (IQ) of a 4th ESO class on January 1.

That is, this type of design is based on collecting data at a specific time (at a single time). Its objective is to describe and analyze a series of variables.

2. Longitudinal designs

This second type of designs, the longitudinal ones, study how certain variables (or just one) evolve or change in a group of subjects (or more). That is, they study these variables at different moments in time. For example, in January, February and March (although it could be with time intervals of years, too, or more).

They can also be applied individually for unique cases. The objective is to study the change that occurs in "X" period of time.