You probably noticed that the less you sleep, the more tired you are. You may have also observed that the more you rehearse a skill like writing, the better you get at it. These are simple observations in life that set the foundations of correlational research. Although these observations need to be tested scientifically for them to become facts, these examples are the basis of correlational studies.
- In this explanation, you will find a presentation of correlational studies in psychology.
- The different types of correlational studies will be presented.
- Moving on, you will learn about interpreting correlational studies' results.
- You will also learn why correlational studies do not let researchers establish cause and effect.
- Lastly, the correlational study advantages and disadvantages of psychology will be discussed.
Correlational Study Psychology
Correlational analyses are widely used in psychological research. Correlation research is based on observations between variables; this means there is no experimental manipulation involved.
Correlational research aims to observe whether or not two variables are related and, if so, how strong the association is.
Correlational studies are a non-experimental research method and a statistical analysis used to understand the linear relationship or association between two variables.
The steps that researchers take when designing a correlational study are the following:
- Stating the research question.
- Identifying the variables.
- Writing of hypothesis statements.
- Conducting the research and gathering data.
- Analysing the data.
Types of Correlational Studies
Three types of correlation studies exist, and we will describe them in detail below, with examples. Further, the different study types will be evaluated, presenting the strengths and weaknesses of each one.
Correlational Studies: Naturalistic observation
In naturalistic observation correlation studies, researchers record observations of variables in a natural setting; this is a non-experimental method in which no variables are manipulated.
An example of this type of correlational research is researchers going to a supermarket (natural setting) and observing how many people buy ice cream on a hot day.
A strength of naturalistic observational research is that it allows researchers to observe participants in a natural setting. This makes it more likely that participants will show their real behaviour, increasing the results' validity. In laboratory settings, for example, participants may not behave as genuinely due to the setting itself.
However, some limitations should be considered, such as the difficulty in limiting confounding factors, which can affect and reduce the study's validity.
Correlational Studies: Survey method
The survey method uses surveys and questionnaires to measure the researchers' variables.
An example would be using questionnaires to determine the highest level of education and socioeconomic status.
The research aim may be to determine if there is a relationship between the level of education and the individual's income.
The advantages of this research method are that it is relatively inexpensive, does not take too much time, and can recruit many participants in a short time. The method usually uses random samples for recruitment, so the research results are more generalisable than other sampling methods.
However, respondents may answer in a socially desirable manner rather than honestly, which reduces the validity of the results.
Correlational Studies: Archival research
Archival research is a type of correlational research that uses secondary data, such as previous research, case studies, historical documents, and medical registries, to measure variables.
Using the Children's Health Foundation Paediatric Asthma Registry to observe the relationship between asthma and prevalence in children is an example of archival research.
The advantage of correlational archival research is that it can be cheaper than alternative methods. Data is readily available, and researchers can obtain data that may no longer be collected, such as documents from historical periods.
Nevertheless, the disadvantages of archival research should be considered. While conducting archival research, the researcher has no control over data collection methods, making it difficult to determine if the data is reliable and valid. Another issue is some data may be missing that is needed for the research.
Correlational Studies: Interpretations
In the statistical analysis of correlation data, a correlation coefficient is calculated.
The correlation coefficient (r) is a measure that determines the strength of the relationship between the two variables.
The correlation coefficient (r) values can range from +1 to -1.
A positive number indicates a positive relationship between the variables; if one variable increases, the other is also expected to increase.
A negative coefficient indicates a negative relationship between the variables. If one variable increases, the other is expected to decrease.
A coefficient of 0 indicates no relationship between the two variables.
The value of the correlation coefficient determines the strength of the correlation data:
- When r = 0, then there is no correlation.
- When r is between 0.1- 0.39, there is a weak correlation.
- When r is between 0.4 - 0.69, there is a moderate correlation.
- When r is between 0.7 and 0.99, there is a strong correlation.
- When r equals 1, then there is a perfect correlation.
Scatter plots are typically used to show the relationship between variables by plotting the data when reporting correlation data. Scatterplots allow us to visually see the strength of the correlation and the direction between the variables.
If the data points are close to the gradient line and have a positive gradient, this indicates a positive relationship. If the gradient is negative, the association is negative.
Fig. 1. The scatterplot shows a positive correlation between the two variables.
Correlational Study Cause and Effect
One of the main ideas researchers need to remember when conducting correlational research is that researchers can't infer causation in correlational studies.
Let's say that a research group tests whether there is a relationship between autism and organic food sales. To test this, they gather existing data from governmental databases. And indeed, they find that in the last ten years, autism diagnosis has increased, and so have organic food sales. There is a positive relationship between the variables.
The research does not imply that autism diagnosis makes people buy organic food, nor does it mean that organic food sales cause autism. In this example, it may be obvious, but in real research, researchers need to be careful about making such inferences.
It is possible that, in some cases, one variable does indeed cause the other one. Further experimental research needs to be conducted to support or disprove it in such cases.
Example of Correlational Research
Researching the relationship between variables has been in the spotlight of psychological research for decades.
Examples include studies looking into the relationship between alcohol consumption and unemployment, the relationship between academic performance and career success, or the relationship between income levels and crime.
A correlation study will start by defining the research question. For example, a study may examine the relationship between self-esteem and social anxiety. Based on previous findings, researchers may hypothesise that there is an existing negative correlation between the two.
The negative correlation would suggest that as self-esteem increases, social-anxiety decreases, or vice versa.
Researchers then decide which inventories or questionnaires will be used to measure the two variables. After this, the correlational statistical test will be calculated.
The statistical analysis may provide a significant result in which the correlation coefficient is -0.78, allowing the researchers to conclude that there is indeed a negative association between self-esteem and social anxiety.
An important thing to note in correlational research is that a negative correlation means a specific variable will increase/ decrease. Any of the variables can increase or decrease. The only thing we can be sure of is that as one increases, the other will decrease.
The researchers may plot their data on a scatterplot, so they and readers can visualise the results.
Regarding the causality effect, it is tempting to suggest that low self-esteem makes individuals experience social anxiety. And although this could be the case, it cannot be established with a correlational test.
Fig. 2. An Example of a scatterplot showing a negative correlation.
Correlational Study Advantages and Disadvantages Psychology
In this section, correlational studies' advantages and disadvantages are critically reviewed.
One of the main advantages of correlational research is that it is quick and easy to conduct. It does not require great statistical knowledge for researchers to be able to use it.
Furthermore, correlations can be tested for existing data, which can inspire future research and be helpful when the researcher may have limited access to the phenomenon, e.g. if it's based on past events.
One of the main disadvantages of correlational research is it can't establish whether variables are causally related.
Cause and effect mean that although research can establish a relationship between two variables, it cannot infer whether one of the variables causes a change in the other or vice versa.
Since correlational studies only measure the co-variables, other potential confounding factors are not considered. The confounding variables may be a better explanatory factor for the study's outcome, making it difficult to determine the validity of the results.
Correlational Studies - Key takeaways
- Correlation studies are a non-experimental research method used to understand the linear relationship/ association between two variables.
- Three types of correlational studies are naturalistic observational studies, surveys, and archival correlational studies.
- In the statistical analysis of correlational data, a correlation coefficient is calculated; it tells researchers about the strength and direction of a relationship between two variables.
- The calculated correlation coefficient value can range from -1 to +1.
- Correlation research has many uses in psychology, for example, to obtain preliminary results that inform researchers whether variables should be explored using experimental research to establish causation relationships.
Explanations 
Exams 
Magazine 
