Correlational study is study variation which goal is to reveal any relationship existing between two variables without explaining it's existence or showing cause of chain effect (Gravetter, Forzano, 2018). This study is not based on any controls, manipulations, or interfering with any variables. Only one thing that this study is based on is observation, and “connecting the dots”. It takes into consideration two variables of one individual, or of group of an individuals, and measures them and wants to find out if they are related. It also can be done by estimating more than two variables, then we are dealing with other forms of correlational study (Gravetter, Forzano, 2018).
Use cases for correlational studies
There are specific reasons that lead scientists to use correlational studies as their method of research. The first reason to use correlational study is when researchers suspect that there is no causal relationship between variables. Using correlation takes less effort and is cheaper than designing an experiment, so in this case, choosing correlation as a method is a better option. The second reason scientists conduct correlational studies is when they might suspect the existence of causal relation between the variables, but are unable to use experimental design. The reasons may be, for example: no physical possibility of carrying out the experiment, using correlation being the most practical option, carrying out the experiment would be unethical (Coolican, 2017).
Correlational study types
There are a few basic types of correlation being observed in research results (Blalock, 2018):
- Positive correlation - when the value of one variable rises up, the value of the second variable rises up as well.
- Negative correlation - when the value of one variable rises up, the value of the second variable falls down.
- No correlation - the values of the two variables are totally independent of each other.
The methodology of correlational study
One of the main instruments used in correlational research, for example in psychology, are questionnaires. Questionnaire is a set of questions specifically designed to measure chosen variables. Those variables may include personality traits, behavioral patterns, beliefs and other. The researchers may choose to measure correlation between those variables and the results of another questionnaire or other variable, e.g. demographic data.
- Armstrong, J. S. (Ed.). (2001). Principles of forecasting: a handbook for researchers and practitioners, Springer Science & Business Media, (30).
- Blalock Jr, H. M. (2018). Causal inferences in nonexperimental research, UNC Press Books.
- Coolican, H. (2017). Research methods and statistics in psychology, Psychology Press.
- Gauch Jr, H. G., Gauch, H. G., & Gauch Jr, H. G. (2003). Scientific method in practice, Cambridge University Press.
- Gravetter, F. J., & Forzano, L. A. B. (2018). Research methods for the behavioral sciences. Cengage Learning.
- Gray, C. W. (1976). The concept of dichotomous correlation. Scandinavian Journal of Psychology, 17(1), 153-159.
Author: Mateusz Fudala