Correlational research is a type of non-experimental research in which investigators measure two variables and establish a statistical relationship between them (correlation) without having to include external variables to reach relevant conclusions. For example, the correlation between time spent studying a subject and grades obtained can be investigated.
There are two essential reasons why researchers are interested in these statistical relationships between variables and are motivated to conduct correlational investigation.
The first is because they do not believe that the relationship between these variables is accidental, that is, a researcher would apply a survey whose use is known by a previously chosen group of people.
The second reason why this type of research is carried out, and not by experimentation, is because of the causal statistical relationship between the variables, in this way, researchers cannot independently manipulate the variables, since it is impossible, impractical and unethical
There are three types of correlational research (natural observation, surveys and questionnaires, information analysis). Likewise, the correlation between variables can be positive (directly proportional) or negative (inversely proportional). Indicating the way in which one variable can affect the other.
It is generally believed that correlational research should involve two quantitative variables such as scores, results of the number of repeated events over a period of time.
However, the most important feature of correlational research is that the two treated variables are measured (without being manipulated) and the results are true, regardless of the type of variable (quantitative or categorical).
Characteristics and definition of Correlational research
The term correlation is defined as the relationship between two variables. The main purpose of using correlations in the field of research is to find out which variables are connected to each other. In this way, a specific event is understood scientifically as a variable.
Correlational research consists of looking for different variables that interact with each other, in this way, when the change in one of them is evidenced, it can be assumed how the change in the other directly related to it will be.
This process requires the researcher to use variables that he cannot control. Thus, a researcher may be interested in studying a variable A and its relationship and impact on a variable B.
For example, a researcher could study the preferred type of ice cream according to age, identifying consumers’ preferences based on their age. The world is full of correlated events, where if variable A is affected, there is a high probability of that variable B is also affected.
Within correlational research, there are two different types, one positive and one negative. Positive correlations mean that variable A increases and, consequently, variable B. On the other hand, when talking about negative correlations, when variable S increases, variable B decreases.
Correlational research is based on various statistical tests that indicate correlation coefficients between variables. These coefficients are represented numerically to indicate the strength and direction of a relationship.
types of Correlational research
Within the correlational research process, the researcher does not always have the opportunity to choose the variables he wants to study. When this happens, it is said that semi-experimental research is being carried out.
There are three types of correlational research within which variables may or may not be controlled. This depends on what kind of approach you take to a particular topic and how you want to conduct the investigation.
1- Natural observation
The first of the three types of correlational research is natural observation. In this way, the researcher observes and records the variables in a natural environment, without interfering with their course.
An example of this could be a classroom. The researcher can analyze the results and the final grades obtained by the students in relation to their level of absenteeism.
This type of correlational research can be time consuming and does not always allow for control over variables.
2- Surveys and questionnaires
Another type of correlational research occurs when surveys and questionnaires are carried out from which information is collected. Within this type of research, a random sample or group of participants must be chosen
For example, when a survey is satisfactorily completed about a new product in a shopping mall, a survey is performed for correlation purposes. This type of research is used to predict whether or not a product will be successful.
Using lookups in correlational searches is often highly convenient; however, if participants are not honest, they can alter the final survey results in a number of ways.
3- Information analysis
The last type of correlational research that can be performed is to analyze data previously collected by other researchers. For example, you can check a population’s court record to predict how crime statistics influence the local economy.
Query files are often freely available as query tools. However, to establish a meaningful correlational relationship, it is often necessary to have access to large amounts of information.
In this type of research, researchers have no control over the type of information that was recorded.
Examples of Correlational research
ice cream truck
A good way to explain how correlational research works would be to think of an ice cream truck. In this way, a person can learn to recognize the particular sound of an ice cream truck, being able to perceive it from a distance.
When the sound of the truck gets louder, the person can recognize that the truck is closer.
In this way, variable A would be the sound of the truck and variable B would be the distance the truck is located. In this example, the correlation is positive, the one that, as the sound of the truck increases, the closer it will be.
If we had different truck sounds, an individual would be able to recognize them all and relate them to different variables.
Identifying autism in children
In this investigation, a study group was used with a test designed to identify differences between different population groups, with the aim of determining whether there was any correlation between the analyzed variables.
A sample of 66 participants, all 12-month-old children, was performed. Among these participants, 35 children had older siblings with a clinical diagnosis of autism. The remaining 31 children had siblings who did not have any degree of autism.
All participants were asked to manipulate an object to perform a specific task and thus be able to identify some kind of normal and abnormal behavior.
At 24 or 36 months of age, the same group of children were again analyzed to determine whether they tended toward autism or developing developmental problems.
The results indicated that 9 of the babies who had autistic siblings were also diagnosed with some degree of autism. A series of correlations for these children were computed, including their scores on the initial manipulation test and the subsequent test.
We could see how the atypical handling of an object by a 12-month-old child correlated positively with the subsequent diagnosis of autism. Likewise, it was negatively correlated with topical or normal child development.
Racism in African American women
In this investigation, three initial questions were raised about the experiences that African-American women may have had in the past.
These questions inquired about the proportions in which these women had experienced some form of racism.
Questioning, therefore, the relationship of these experiences with possible psychological conditions of women and the ability of these women to mitigate the impact that racism had on their psychological condition.
The sample included 314 African-American women who responded to a written survey designed to measure their experience with racism, the possible psychological conditions derived from this phenomenon and the choice of behaviors to deal with situations of discrimination.
The results indicated the manifestation of numerous forms of racism (insults from co-workers, being ignored by salespeople in department stores, racist jokes, among others).
These different forms of racism were reported by more than 70% of the participants. It was evident that racism was a common experience among African-American women.
Correlational coefficients revealed a significant positive relationship between reported racism and the events and possible psychological problems of these women. This conclusion covered the mechanisms for dealing with racism employed by them.
Other results indicated that the different models used by African American women to deal with these incidents were regularly employed with partial success.
In this way, many women would have turned a negative experience into a much worse experience in an attempt to mitigate its psychological impact (Goodwin & Goodwin, 2017).
