What is research design and how is it done?

We analyze the set of tools, techniques and methods suitable for conducting research.

Do you know what a research design is? Surely you relate it to statistics classes, experimental psychology... However, it is a tool that is used in many types of research and in many fields of study.

It is a set of techniques that allow a researcher to prepare his experiment or research. In this article we will learn exactly what it consists of, what types exist and how some variables influence its design.

• We recommend reading: "The 15 types of research (and characteristics)".

What is research design?

Before explaining how these designs work and what types exist, let's see what a research design consists of, or how it is defined. It is a set of techniques and methods chosen by a researcher to carry out an experiment or a research project.. It is a series of steps that serve as a guide to the researcher.

Thus, in other words, the research design is a structured and specific plan of action, aimed at the design and implementation of an experiment. It encompasses a set of rules or concrete steps that make it possible to reach the researcher's objective, in this case, to develop an experiment (whether observational, experimental, quasi-experimental...).

In other words, they make it possible to create research projects; they establish the necessary steps to reach them.

Scientific method

The research design is framed within the scientific method, which consists of a general procedure characterized by systematization and control. The scientific method (and therefore the research design) is divided into two: quantitative and qualitative.

1. Quantitative design

In the quantitative design or method, phenomena are events determined by causes independent of human will, and are therefore objective. In this case, this type of research can be carried out when it is important for the researcher to obtain statistical conclusions that help him to gather information for his research. That is, it is based on calculation and numbers.

• Recommended: "The 9 differences between qualitative and quantitative research".

2. Qualitative design

In the qualitative design, on the other hand, reality is considered as a subjective construction, in which many constructions of the same reality are possible; moreover, in this case the subject and the object of knowledge are interdependent.

Furthermore, qualitative research makes it possible to establish relationships between the data collected; on the other hand, the different theories related to natural phenomena can be tested (refuted or proven) through this design and the corresponding mathematical calculations. Often these types of designs are intended to answer open questions, such as: "why", "how", "why", "why", "why", "why", "why", "why" and "why".

Methodology

What methodology is used to develop a research design? It can be of different types: deductive, inductive and hypothetico-deductive. The deductive methodology is the one that goes from the general to the particular (emphasizes the theory); the inductive, the one that goes from the particular to the general (emphasizes the data), and finally the hypothetico-deductive is the one that combines the two previous ones.

Degree of control

In addition, depending on the degree of control we want to have in our experiment, the research design can be of 4 types. These are detailed below.

Experimental design

The experimental research design is the one that has the maximum degree of control (by the researcher); that is, there is manipulation of the variables. In addition, it allows the establishment of causal relationships between variables.

2. Comparative design

The comparative design, in turn, is divided into two more: the correlational (when there is a degree of relationship between the variables; it does not allow causal relationships to be established) and the comparative design (where the independent variable is a selection variable; that is, the subject comes with its "set" value [for example, race or gender]).

On the other hand, through the properly comparative design, quasi-causality relationships can be established. In both types of comparative designs the degree of control by the experimenter is intermediate.

3. Observational/survey design

This type of research design has a minimal degree of control by the researcher; that is, there is no manipulation, it is simply observed. As in the case of the correlational design, this type of design does not allow causal relationships to be established between variables.

An example of observational research design is surveys.

Type of manipulation

Closely related to the previous section, we ask ourselves: What type of manipulation can be performed in a research design?

Logically, it will depend on the type of design and experiment. Basically, there are three types of research designs depending on the type of manipulation applied to the independent variables.

Experimental research

This first research design involves an intentional manipulation of the independent variables. In addition, there is randomization of the independent variables.

2. Quasi-experimental Research

In this second type, the manipulation is selection, not intentional (i.e., it involves variables or values already possessed by the subject; for example, sex). In this case, there is no randomization of the variables (in fact, this is the main difference between an experimental and a quasi-experimental research design).

3. Non-experimental research

Finally, in non-experimental research there is no manipulation of variables or randomization of variables.

Variables in research

Another important concept that we must know in order to understand well what the research design is, are the variables in research.What do they consist of?

A psychological variable is a psychological construct that can be measured with mutually exclusive values (e.g. gender, level of anxiety, degree of well-being, weight, height, etc.). Variables can be of different types (according to different classifications). One of the most important is the one that divides them into:

1. Dependent variables.

The dependent variable, usually expressed by "AND", is the effect produced by the independent variable. For example, it can be the degree of anxiety (which increases or decreases as a function of a treatment).

Independent variables

The independent variables, however, are represented by "X", and are the cause of the effects. That is, following the previous example, it would be the psychological treatments (independent variable), for example, that influence the degree of anxiety (dependent variable).

Meta-analysis

Finally, another interesting concept to know in relation to research designs is meta-analysis; this is a technique for evaluating the quantitative results of a set of empirical studies. Through it, a summary or synthesis of the significant effects of the variables studied is reached.

This is a type of methodology consisting of the systematic review of different studies and results; through it, a series of statistical techniques are applied to analyze these results, mainly quantitatively.