The independent variable and the dependent variable they form two of the best-known categories in the world of science and research in fields such as medicine, psychology, sociology and other fields of knowledge.
These are not just basic concepts for conducting experiments; in addition, they help to understand the functioning of reality from the analysis of isolated phenomena. In short, they allow us to reduce the complexity of what we have studied and focus on simple elements that can reveal scientific knowledge.
In this article, we will see what are the dependent and independent variables, with several examples to help understand its role in science and in the use of statistical tools.
Dependent and independent variable: what are they?
In psychology, as in any other scientific discipline, research is essential to lead to the development of new techniques, methods, explanatory models and practical applications, or to improve or ensure the safety and veracity of pre-existing ones.
And to study something, we have to keep in mind that in any experiment it is necessary to evaluate and manipulate different variables. Variables are traits or characteristics that can vary by adopting different values or categories, and the variation can give us clues as to how this occurs or why a phenomenon appears that we are interested in studying.
The variables are therefore elements of reality that we can define in a specific and predictable way to the point that we find repeated several times in nature or in society what it refers to. For example, gender is a variable, and what it indicates is reflected in most of the human beings we observe, with very few situations that present any ambiguity.
At the operational level, whenever we work experimentally, we will do it with two main types of them: dependent and independent variable. Let’s examine each of them throughout this article.
Basic definition of independent variable
It is defined as a variable independent of any variable tested at the experimental level, manipulated by researchers in order to test a hypothesis. It is a property, quality, characteristic or ability that has the power to affect other variables, Be able to modify or mark the behavior of the rest of the variables.
Thus, the different values of this variable will be fundamental to design and interpret the results of the experiment, as can be explained.
For example, you can mark the different situations that participants will go through during the experiment (if more than one occurs) or the groups that will go through different experimental conditions. In these cases, we could speak respectively of independent intrasubject or intersubject variables.
The independent variable sand is so called precisely because its values will not be changed by the other variables of the experiment itself.. Gender or age are generally independent variables, as they do not change based on a few variables. Of course, we can use them to study other variables.
In all cases, the variables are dependent or independent depending on the context in which we find ourselves. In one research, the preferred musical genre may be the dependent variable, and in another, it may be the independent variable.
Dependent variable: concept
As for the dependent variable, we speak of this quality or characteristic behavior is affected by the independent variable. This is the variable (s) measured in order to interpret the results. In other words, it is what is observed to see if it changes, or how it changes, if certain conditions are met (controlled by the use of dependent variables).
In this way, we are confronted with the type of variable that we analyze in experience or research, evaluating its behavior according to the values of the independent. If the independent variable is the cause, we could think of the dependent variable as the effect we are measuring and the fact that we have manipulated the first.
Of course, we have to consider that not all research using dependent and independent variables express causal relationships. In other words, the fact that changing the value of the independent variable also changes the value of the dependent person in a more or less predictable pattern does not mean that the cause of the latter change was the manipulation of the independent variable. Especially in the social sciences, such phenomena can express a simple correlation effect.
For example, if asking for the intention to vote for those with a lower level of education gives a different result than asking for the intention to vote for those with a university degree, this should not mean that the independent variable “level of education” is the one that generates this variation; there may be another hidden variable that explains both the different voting intentions and the low level of education, such as lack of financial resources.
Details on its use in research
The division between dependent variable and independent variable is a basic element that is part of any research carried out. But the number of variables to consider, as well as the type of experimental design and what is actually meant to be analyzed, can vary widely.
For example, a simple design may only require the use of one independent variable and one independent variable. In general, it is generally recommended that at least when it comes to the independent variable, we only use one at a time, as the larger the number of freelancers, the more complex the experience and the possibility to cause measurement error is large.
However, if, for example, we want to assess the effects of a drug, it is more appropriate to assess different elements in the same experiment. We could have an independent intergroup variable, which would be the type of group (group of subjects with drug and group of control subjects, in order to see if there are significant differences) and an intra-group which would be the time of treatment ( preprocessing, postprocessing and monitoring).
Moreover, as dependent variables, we could assess different aspects such as levels of depression, suicidal thoughts, eating habits, libido, quantity and quality of sleep.
In any case, the relationship between the dependent and independent variables will be the same and you should always check if there is an effect of each of the independents on the dependents (and not only of each independent but also if the interaction between those – this has a dependents effect). This can be assessed through different types of design, such as ANOVA.
Another aspect to consider is that depending on what you are going to study and how this research is going to be conducted, the same reality can be variable or independent.
For example, a person’s body mass index can be an independent variable if it is used to assess whether it affects another variable, or it can be a dependent variable if we feel that the same BMI can depend on ‘another variable. It is therefore rather the position from which we analyze the variable than the variable itself that makes it dependent or independent.
Examples of its use in science
In conclusion, we see some examples of situations or research where we can see a dependent variable and an independent variable.
A first case could be a study aimed at analyze the level of change in heart rate that generates exposure to different height levels in people with acrophobia. In this case, the height at which the subject is exposed would be the independent variable, while heart rate would be a dependent variable.
Another study could be to analyze the effects that the type of language used in self-esteem assessment instruments can have on patients’ self-report. Language type can be an independent variable and the results of self-esteem questionnaires depend on it.
A third example could be research that analyzes the effect of sedentary / physical activity levels on body mass index, BMI being the dependent variable and physical activity levels independent.
A fourth and final example can be found in a study that assesses how a positive effect affects levels of life satisfaction. The levels of positive affect would be the independent variable, and the dependent variable would be the levels of vital satisfaction.
- Franklin, J. (2002). The science of conjectures: proofs and probabilities before Pascal. Baltimore: Johns Hopkins Univ Press.
- Donat, LM (2008). The SAGE Encyclopedia of Qualitative Research Methods. Los Angeles: SAGE Publications.
- Piracy, I. (1990). The domestication of chance. Cambridge: Cambridge University Press.
- Lakshmikantham, V., Kannan, D. (2002). Manual of analysis and stochastic applications. New York: M. Dekker.
- Mesly, O. (2015). Creation of models in psychological research. New York: Springer Psychology.
- Michell, J. (1986). Measurement scales and statistics: a clash of paradigms. Psychological bulletin. 100 (3): 398-407.