Types of Variables – Definition & Examples

10.12.22 Types of variables Time to read: 10min

How do you like this article?

0 Reviews


types-of-variables-01

A fundamental component in statistical tests is the methodology you employ in selecting your research variables. The careful selection of appropriate variable types can significantly enhance the robustness of your experimental design. This article explores the diverse types of variables and their classification, accentuated with various examples.

Types of Variables – In a Nutshell

A variable is an attribution of a specific piece of data to an item in research. It is used to enable analysis and evaluation of studies in science, economics and other areas.

Definition: Types of variables

A variable is a trait of an item used for analysis in research. Types of variables in research are imperative, as they describe and measure places, people, ideas, or other research objects. There are many types of variables in research. Therefore, you must choose the right types of variables in research for your study.

Variables can be differentiated into several categories. One way is to separate independent from dependent variables, referring to whether it is the cause or effect of something. Furthermore, types of variables can be distinguished as qualitative or quantitative. Qualitative variables can then again be defined as nominal or ordinal, while quantitative ones get separated as discrete and continuous. And eventually, there are special types of variables that belong to one of those categories.

Conduct a final format revision for a print of your thesis
Before submitting your thesis for print, check on your formatting with our 3D preview function for a final time. It provides an exact virtual visualization of what the printed version will resemble, making sure the physical version meets your expectations.

Independent vs. dependent variables

The purpose of experiments is to determine how the variables affect each other. Therefore, you need variables you manipulate, those you keep steady to not accidentally change the conditions and those, on which you measure the effects on, called independent vs. dependent variables.

The following examples are based on a theoretical research project, which wants to investigate how the concentration of salt in water impacts the growth and survival of plants.

Types-of-variables-independent-dependent

The independent variable is the one the researcher manipulates. There are various synonyms for this type of variable, including treatment variable or explanatory variable. However, the latter one mostly refers to conditions the researcher themselves cannot manually manipulate, such as the weather. If you would like to know more about explanatory vs. response variables, please follow the link to the corresponding article.

Example

The independent variable would be the amount of salt that is dissolved in the water. The concentration of salt will be manipulated by the researcher to see how it changes the effects on the plants.

The dependent variable shows the effects of the manipulation. It is also called a response variable, as it consists of the response to the changes of the independent variable.

Example

The dependent variable would be the growth of the plant in this experiment. The researcher waters the plant with a specific salt concentration and then measures what effect this has on the growth and survival of the plants.

To make sure that the observed results are truly related to the independent variable, it is necessary to keep every other factor stable and consistent to avoid research bias. These steady factors are called control variables.

Example

Control variables in the plant growing experiment would be the amount of water given, the temperature where they are stored, the lighting, etc.

The table below summarizes all three types of variables.

Types of variables in research Explanation Example (based on our example)
Independent variables The variables you manipulate to affect the experiment outcome The amount of salt added to the water
Dependent variables The variable that represents the experiment outcomes The plant’s growth or survival
Control variables Variables held constant throughout the study Temperature or light in the experiment room

Mediator vs. moderator variables

In an experiment, however, there are more than these three types of variables to be considered. Important influential factors are also mediator vs. moderator variables, which have different effects on the results of a study.

Types-of-variables-mediator-moderator

Mediator variable

A mediator variable is one that explains the connection between the independent variable and the dependent variable. It is, to say it simply, the explanation for why the manipulation affects the result.

Example

In an academic context, the mediator between study time as an independent variable and test results as a dependent one could be the confidence in your abilities before and during the test. When you studied a lot, you are sure of your capabilities and thus most likely perform better in the exam as you are calm and trusting.

Moderator variable

A moderator variable, on the other hand, influences the manifestation of the results, whether they get stronger or weaker.

Example

For example, customer loyalty to a brand causes more stable sales for a company. The customer loyalty, however, varies in different age groups, as younger people tend to be not as committed to one brand as older people are. This way, the age of the target group moderates the strength of connection between the two variables.

Extraneous and confounding variables

These two types of variables can involuntarily influence your study and therefore need to be eliminated. To be aware of potential bias, it is essential to look out for these.

  • An extraneous variable is one, which is not considered in the experiment, but still has an influence. This can easily lead to research bias and warped results of the study. Therefore, it is important to know about the exact circumstances of your experiment and make sure that everything except for the independent variable stays consistent.

Examples

  • Does a lack of sleep influence exam performance? Extraneous variables could be: Test anxiety, importance of the exam, affection towards the subject of the test, …
  • Your company launches a new product, and you want to test how intuitive the usage is. Extraneous factors could be: user experience with similar products, ability to understand the manual, …
  • A confounding variable is a type of extraneous variable, which is related to the independent variable. This means that a factor apart from the independent one is the actual cause for a result.

Example

  • Your study is about how regular exercises and weight loss are correlating. Your confounding variable, however, is a change in the diet of your participants. The regular exercising caused them to think about their health more, changing their eating habits, which are actually the cause of the weight loss.

Differentiating in independent, dependent, and control variables only explains what role a variable has in an experiment. To distinguish the actual types of variables, their appearances, and rules of use, we need to separate them according to different criteria. However, it is important to know that all three types above can, depending on the definition of the individual case, be sorted into the categories of the next paragraphs.

Types of variables

If you want to classify the types of variables, you can sort them into different sets, starting with the primary categories of qualitative and quantitative variables. These can then furthermore divide into subsets to specify their characteristics.

Types-of-variables-qualitative-quantitative

Qualitative/Categorical variables

Qualitative variables, also called categorical variables, are those represented by names or scales. They can, in some cases, also be defined by numbers. However, these numbers cannot be used in statistical calculations because they are merely simplified replacements for words. Categorical variables can be further divided into nominal and ordinal ones.

Nominal variables

Nominal variables are always single choice questions, where you usually have to select one out of two or more options. Another application would be forming categories for various purposes. The essential aspect is that the different variables do not have an order to sort them by. The simplest nominal variable is the binary variable, which consists of only two options to choose from.

Examples

  • Married or single, minor or adult, Christian or Muslim or Jude or others, …
  • Categories in an online shop: Shoes, skirts, trousers, shirts, …

Ordinal variables

Ordinal variables, on the other hand, always pose a ranking of some sort, which can either consist of numbers or labels. The important point is here that even if you use numbers for ranking, they are not empirically measurable.

Examples

  • How happy do you feel? Very happy – happy – medium – unhappy – very unhappy
  • How satisfied are you about this survey, on a scale of 1-10?

Quantitative/Numerical variables

Numerical or quantitative variables hold a real value and have a mathematical meaning. Contrary to qualitative variables, these can be of statistical use in calculations. This type of variable can be differentiated into discrete and continuous ones.

Discrete variables

Discrete variables always have a distinct value, which mostly refers to whole numbers. Usually, they count occurrences, people, actions and other things that are countable.

Examples

  • You can count the participants with red hair, for example, meaning that there cannot possibly be 2.5 red haired participants.
  • You can analyse the number of times a person buys something from a specific store.

Continuous variables

Continuous variables can take any value, such as in height, money in a bank account, time, or distances.

  • Interval variables focus on the discrepancy between two data points, meaning that the margin of both is of interest. Another important criterion is that with this type of variable, 0 is a valid reference point, which can be the value of a result.

Examples

  • The difference in temperature from summer and winter.
  • The comparison of credit points you earned in two takes of an exam. (With 0 points meaning no academic abilities and a difference of 0 meaning an equality in performance)
  • Ratio variables, on the other hand, have an absolute zero value, meaning that 0 marks the absence of existence and data.

Examples

  • The age of a person, which can never be zero because it has a minimum of a few seconds.
  • If you compare two data sets, the value of the third is 0 because it does not exist.
Ensure Originality in Your Thesis
Double-check that your thesis is free from unintentional plagiarism!
In only 10 minutes, you can receive your easy-to-follow plagiarism report using the BachelorPrint Plagiarism Checker. All citation errors in your thesis will be detected and listed by the plagiarism software.

Special types of variables

Variables can be chosen in almost every way to suit your research, regardless of which or how many you need. There are, however, some predefined special types of variables, which will be explained in the following.

Random variables are those, whose result is based on a random event. They typically have a set of possible outcomes, out of which one of them coincidentally occurs. Mostly, they are discrete, but can in some cases also be continuous

Examples

  • Discrete: rolling dice
  • Continuous: price of a product in a random store within a price range

The algebraic variables in terms are the most familiar one to students and mostly defined as x. It is used in mathematics as a space holder for an unknown value.

Examples

  • 2+x=5 → x=3
  • x*7=56 → x=8

Binary variables are those who only have two outcomes. They can be sorted into the class of nominal variables, as the two possible results do not hold any mathematical meaning.

Examples

  • Coin toss
  • Success or failure
  • Yes or no

Latent variables cannot be measured directly, but instead need other observable variables to quantify them. They often refer to immeasurable things like satisfaction, wellbeing, or stress, where numbers cannot give exact results. You can approximate their value through ordinal values, but this is only a self-assessment of the test subject and might not be veracious.

Example

An example could be a customer satisfaction survey. While you cannot measure satisfaction as a concept, you can approximate it through more detailed questions. In an online shop, you could measure the time they spent on the website, how many products were bought, as well as asking whether they would consider buying there again, etc.

When variables are highly related to one another, you can consider merging them into one composite variable. In many cases, composite variables are also latent variables.

Example

For example, customer satisfaction when you know that the number of products bought, the money spent and the time they stayed on the website are highly related, you can combine them into the composite variable of customer satisfaction. If, however, there appears to be an observed correlation between the amount of money spent on clothes and the times they booked a holiday trip to Alaska, they can most likely not be combined as there is no realistic correlation.

Examples

The following table will give you an example for each variable in the context of academic activities.

Type of variable Example
Independent variable Hours of studying
Dependent variable Result of the exam
Control variable Hours of sleep, sports activities, nutrition
Mediator variable Repetition contributs to memory
Moderator variable Interest in the subject
Extraneous variable Test anxiety and nervousness
Confounding variable Participants studied right before going to bed, leading to better remembrance
Qualitative/Categorical
Nominal variable Did you study? (Yes or no)
Ordinal variable How confident are you to pass on a scale of 1-5?
Quantitative/Numerical
Discrete variable How many times did you attempt to pass this exam already?
Continuous variable
Interval variable The difference in credit points between your first and second attempt.
Ratio variable The time needed to finish the exam, which cannot be 0 minutes.

FAQs

The simple answer is: a lot. In a cause-effect experiment you will have the independent, dependent and control variable, as well as mediator and moderator ones. Differentiating by the characteristics of each variable, they can be divided into qualitative and quantitative ones. Qualitative variables are nominal and ordinal-scale variables, while qualitatives are discrete or continuous. The classification of different types of variables is not always easy, which is why we wrote a whole article with detailed descriptions about each one.

Mediator variables sit between independent and dependent ones and explain their relationship, while moderator variables modify the strength of that relationship or connection.

To define it, you first need to set the type of variable it has to be, give it a name and, if necessary, the values it can take. When doing research, it might also be helpful to define what the variable is not, to avoid confounding and misinterpretation.


From

Salome Stolle

How do you like this article?

0 Reviews
 
About the author

Salome Stolle works as the brand manager for the English market at BachelorPrint. Throughout her 12-year residency in Denmark, she completed her International baccalaureate and Master’s in Culture, Communication, and Globalization with a specialization in media and market consumption. Through this experience, she has gained advanced competencies in academic writing and a high proficiency level in the English language. With her passion for writing, she does not only deliver well-written content but also strives to adjust to the students’ demands.

Show all articles from this author
About
BachelorPrint | The #1 Online Printing Service
For Canadian Students

As a leading expert in the printing and binding of academic papers, theses, and dissertations, the BachelorPrint online printing service furnishes a diverse range of bindings and configuration options. BachelorPrint’s goal is to elicit a proud smile on every single Canadian student’s face, as they hold their ideal binding in their hands.<br/>Moreover, BachelorPrint provides numerous educational articles on insightful subjects related to academic writing in their Study Guide, assisting students during the process of writing their thesis or dissertation.


New articles