Correlational research consists of evaluating two variables, its purpose being to study the degree of correlation between them.
Correlational research, therefore, tries to discover how one variable varies as the other does. However, in this case, we only study the direction of movement and the intensity of the relationship. On the other hand, we must know that correlation does not imply causality. Likewise, to know the degree of variation, it is necessary to calculate some type of regression; like linear or multiple.
Why conduct correlational research?
This type of research follows a protocol based on the scientific method. We ask the questions first. Afterwards, we observe to make a first impression. Next, we measure the variables of interest. Finally, we analyze and draw conclusions.
There are also several reasons why it may be of interest to carry it out:
- First, it allows us to know something as important as the correlation between two or more variables. That is, it tells us how one variable varies when we modify the other. In this way, the possible random effect is ruled out and possible accidental manipulation is avoided.
- It is usually the starting point in regression models. Once we know the degree of variation and the direction of the compared variables, we can generate an explanatory model.
- One of the biggest drawbacks is that it does not allow a cause-effect relationship to be established. To know these relationships, other statistical techniques would have to be carried out and, above all, a review of the existing literature would have to be carried out.
Characteristics of correlational research
It is convenient to know some of its main characteristics, which would be the following:
- It is based on the previous descriptive analyzes of the information. In this way, once we know the measures of each variable, we can study their relationships.
- It allows studying the relationship between variables without the need to manipulate them.
- Provides information based on comparable values.
- It allows us to know the correlation between two variables. That is, how one varies when another is modified. In addition, it informs of the direction of said variations.
- The main statistic used to know the degree of relationship between two variables is the linear correlation coefficient for quantitative variables.
- The variant of the Spearman coefficient is used in the case of nominal or ordinal variables. Both allow us to know the degree of correlation.
Correlational Research Example
Let's imagine that we have certain data on students of the degree in economics. We carry out a preliminary documentary investigation and uncover relevant information. There seems to be a relationship between grades and variables such as parental income. To study it, we decided to conduct a survey and income is classified into three levels (ordinal variable).
We can observe that the process is similar to that of other types such as the experimental one. First we need to know what we are looking for, the relationship between variables. Later, how we will study it, in this case using the Spearman coefficient. We then apply it and analyze the information obtained. The last step is to establish the conclusions.
Tags: banking Spain famous-phrases