The Different Types Of Variables Used In Research And Statistics

Scientists and statisticians conduct experiments on a regular basis. Scientists use these experiments to identify cause and effect, while Statisticians use variables to represent the unknown or varied data in their experiments.

Determining which variables to use is vital to the experiment. Also, choosing the right variables will lead to clearer analyses and more accurate results.

What Is a Variable?

A variable is something you can control, manipulate, or measure when conducting research or experiments. They are characteristics, numbers, or quantities and can represent specific items, people, places, or an idea. The variables may be referred to as data items.

The variables in an experiment will vary depending on the desired outcome. All scientific experiments and statistical studies will analyze a variable.

They are referred to as variables because the values can vary. A variable’s value can change within a single experiment. Whether there is a change between the groups being studied or the value changes over time, it may not necessarily be a constant.

Designing Experiments

As noted, all scientific experiments and statistical studies will control, manipulate, or measure a variable. In fact, the experiments are usually designed to determine the effect one variable has on another variable, cause, and effect.

When designing experiments, it is extremely important to choose the right variables. Choosing incorrectly can skew the results and derail the experiment or study completely. Choosing right can help an experiment or study run much more smoothly and produce more accurate results.

It is not just the specific variable within the experiment that needs to be determined, but the variable type as well. Knowing the variable type will allow you to interpret the results of the experiment or study.

It should be noted, though, that categorizing variables is a little subjective. Scientists and statisticians have some wiggle room when they categorize their experiment variables.

Generally, you will need to know what data the variable represents and what part of the experiment the variable represents in order to determine the variable type.

Independent, Dependent, and Control Variables

Typically, there will be an independent variable, dependent variable, and control variable in every experiment or study conducted.

Independent variables. Independent variables are the variables in your experiment that are being manipulated. They are referred to as independent variables due to the fact that their value is independent of other variables, which means that the other variables cannot change the independent variable.

They can, however, change the other variables in your experiment. They are sometimes referred to as an experimental variable or a predictor variable and are the cause in the cause and effect.
Dependent variables. Dependent variables are the variables in your experiment that rely on other variables and can be changed or manipulated by the other variables being measured.

They are the effect in the cause and effect and are sometimes referred to as the outcome variable. Those conducting the experiments will usually attempt to determine what causes this variable to change and how it is affected.
Control variables. Control variables are the variables in your experiment that are constant. They do not change over the course of the experiment or study and will have no direct effect on the other variables being measured.

Control variables are not technically part of the experiment, but they can help ensure more accurate results and actually make it easier to reproduce the experiment or study.

Qualitative Versus Quantitative Variables

Every single variable you include in your experiment will need to be categorized as either a qualitative variable or a quantitative variable.

Qualitative variables. Also referred to as categorical variables, qualitative variables are any variables that hold no numerical value. They are nominal labels. For example, eye color would be a qualitative variable. The data being recorded is not a number but a color.

These variables don’t necessarily measure, but they describe a characteristic of the data set. They can be broken down further as either ordinal variables or nominal variables (see below for definitions).

Other examples of qualitative variables are the city of origin, dog breed, hair color, college major, and car color.
Quantitative variables. Quantitative variables, or numeric variables, are the variables in your experiment that hold a numerical value. Unsurprisingly, they will represent a measurable quantity and will be recorded as a number.

These variables will measure “how many” or “how much” of the data being collected. These can be broken down further as either continuous variables or discrete variables (see below for definitions).

Population, height, GPA, average gallons of gas sold, and the number of dogs owned would all be considered quantitative values.

30 Other Variable Types Used in Experiments

This is by no means a comprehensive list, as the list of all variable types would be difficult to document in one place.

Below are many of the common and some less common variable types used in scientific experiments and statistical studies. Included is a brief overview of what that variable type measures.

Active variable. An active variable is a variable that can be manipulated by those running the experiment.
Antecedent variable. Antecedent variables come before the independent and dependent variables. With “antecedent” meaning “preceding in time or order,” this is not surprising.
Attribute variable. An attribute variable also called a passive variable, is not manipulated during the experiment. It may be a fixed variable or simply a variable that is not manipulated for one experiment but could be for another.
Binary variable. Binary variables only have two values. Typically, this will be represented as a zero or one but can be yes/no or another two-value combination.
Categorical variable. Categorical variables are variables that can be divided into larger buckets or categories. Shoe brands, for instance, could include Nike, Reebok, or Adidas.
Composite variable. This variable type is a bit different from others. A composite variable is made up of two or more other variables. The individual variables that make up the composite variable will be closely related either conceptually or statistically.
Confounding variable. Confounding variables are not good. They can affect both independent and dependent variables and invalidate results. Sometimes referred to as a lurking variable, these variables are considered “extra” and were not accounted for during the designing phase.
Continuous variable. Continuous variables have an infinite number of values between the highest point and lowest point. Distance is a continuous variable.
Covariate variable. A covariate variable can affect the dependent variable in addition to the independent variable. It will not be of interest in the results of the experiment, though.
Criterion variable. This is a statistical variable only. It is another name for the dependent variable.
Dichotomous variable. This is another name for a binary variable. Dichotomous variables will have two values only.
Discrete variable. Discrete variables are the opposite of continuous variables. Where continuous variables have an infinite number of possible values, discrete variables have a finite number.
Endogenous variable. Endogenous variables are dependent on other variables and are used only in statistical studies, in econometrics specifically. The value of these variables is determined by the model.
Exogenous variable. An exogenous variable is the opposite of an endogenous variable. The value of this type of variable is determined outside of the model and will have an impact on other variables within the model.
Explanatory variable. This is a commonly used name for the independent variable or the variable that is being manipulated by those running the experiment.
Grouping variable. A grouping variable is used to sort, or split up, the data set into groups or categories.
Interval variable. Interval variables show the meaningful difference between the two values.
Intervening variable. Intervening variables, or mediator variables, explains the cause, connection, or relationship between two other variables being measured.
Manifest variable. A manifest variable is a variable that can be directly observed or measured within the experiment.
Moderating variable. A moderating variable can affect the relationship between the independent variable and dependent variable. It can either strengthen, diminish, or negate the relationship.
Nominal variable. This is another way of saying categorical value. Nominal values will have two or more categories.
Observed variable. Observed variables are variables that are being measured during the experiment.
Ordinal variable. Ordinal variables are similar to categorical or nominal variables but have a clear ordering of categories. Examples such as High to low and like to dislike would both be ordinal variables.
Polychotomous variable. Polychotomous variables have more than two possible categories or values. These can be either nominal or ordinal.
Ranked variable. Ranked variables are ordinal variables. The researcher may not know the exact value, but they will know the order in which the data points should fall.
Ratio variable. Ratio variables are similar to interval variables but have a clear definition of zero.
Responding variable. Responding variables are the effect or outcome of the experiment. Similar to dependent variables, responding variables will “respond” to changes being made in the experiment.
Scale variable. A scale variable is a variable that has a numeric value that can be ordered with a meaningful metric. It will be the amount or number of something.
Study variable. Often referred to as a research variable, a study variable is any variable used that has some kind of cause and effect relationship.
Test variable. A test variable also referred to as the dependent variable, is a variable that represents the outcome of the experiment.

Never miss an opportunity that’s right for you.

Author

Chris Kolmar

Chris Kolmar is a co-founder of Zippia and the editor-in-chief of the Zippia career advice blog. He has hired over 50 people in his career, been hired five times, and wants to help you land your next job. His research has been featured on the New York Times, Thrillist, VOX, The Atlantic, and a host of local news. More recently, he's been quoted on USA Today, BusinessInsider, and CNBC.

Recent Job Searches

Explore All Jobs