Independent and Dependent Variables
One slide of the supplied teaching source carries the whole of its variables content, and what it offers is a test for telling two variables apart rather than a definition of either. Here is that test, what it is good for, and the eight variable types the material never names.
The whole of what the source says about variables
The material for this topic is one slide, reproduced below in full. It is sourced from a university web page on statistics teaching — the same host the week's study notes draw on elsewhere and mis-credit — and it is the only sustained passage on variables anywhere in the supplied material.
Read as a working test it holds up. Ask yourself what you are trying to influence or predict; that thing is the dependent variable. Ask what you are using to do the influencing or predicting; that is the independent variable. The last sentence gives the relationship a direction — you explain variation in the dependent variable using the independent one.
Four names for two things
THE TWO VARIABLE TYPES AS THE SLIDE RELATES THEM
| Term | How the slide characterises it | Alternative name given | Formal definition supplied |
|---|---|---|---|
| Dependent variable | The one you are trying to influence or predict | "Response" variable | None |
| Independent variable | The one you use to make the predictions, or to determine whether it influences another | "Explanatory" variable | None |
| Variable | Used throughout and never characterised | None | None in this week or the next |
The alternative names are useful in their own right: explanatory and response describe the roles more plainly than independent and dependent do, and both pairs appear in published methods writing.
The direction of the relation is a claim you are making
This is a reading of the source's own wording rather than something it states, and it is worth making explicit. Every clause in the hint is about you: what you are trying to influence, what you are using to make predictions, what you are trying to explain. Nothing in the test inspects the data.
That means the labels do not come out of the world; they come out of your research question. The same measurement can be the dependent variable in one study and the independent variable in another, and the material's own phrasing — "which variable is which in a study" — allows exactly that. Assigning the roles is therefore an assertion about how you think the situation works, made before any data exists to support it, and it is the same assertion a hypothesis makes explicit — see Formulating and Testing a Hypothesis.
What the slide gives you
- A test for telling the two apart, phrased around what you are trying to do
- Two alternative names — explanatory and response — that describe the roles more plainly
- A stated direction: you explain variation in the dependent variable using the independent one
What it leaves to you
- A definition of either term, and of "variable" itself
- Any variable type beyond the two, and any account of what can distort a relation
- Any link to how a variable is measured, coded or analysed later
Applying the hint to a project research question
The three questions below are constructed for this page as illustrations of the test in use. They are not source examples — the slide supplies no worked example of its own — and the relations in them are not findings.
THE HINT APPLIED TO THREE CONSTRUCTED RESEARCH QUESTIONS
| Research question (illustrative) | What you are trying to explain — dependent | What you are explaining it with — independent |
|---|---|---|
| Does the frequency of formal status reporting affect the volume of late change requests? | Volume of late change requests | Frequency of formal status reporting |
| Do teams with longer shared delivery history produce more accurate effort estimates? | Accuracy of effort estimates | Length of shared delivery history |
| Does the seniority of the assigned sponsor predict how quickly escalated decisions are resolved? | Time to resolve escalated decisions | Seniority of the assigned sponsor |
In each row the dependent variable is the one named after "affect", "produce" or "predict". That is a rule of thumb about English sentence construction rather than a source rule — but it is a fast check, because a research question that does not have a direction in its grammar usually does not have one in its design either.
The test also tells you when it does not apply. A question that asks what something is like, or how people experience it, has nothing to influence or predict — and the same week's qualitative approaches are built on exactly that kind of question. The first of the week's five collection steps nevertheless opens with "Define the variables", which is treated on The Data Collection and Analysis Sequence.
The variable types this material never names
Two types are given. The list below is what a reader will meet immediately in any methods text and will not find anywhere in this material — not defined, not exemplified, not mentioned. Treat the lines as signposts for your reading list rather than as definitions; the material supplies neither.
- Categorical, continuous and discrete variables — the vocabulary for what kind of value a variable can take
- Confounding variables — the reason an apparent relation may not be the one you think
- Moderating and mediating variables — how a relation changes strength, or works through something else
- Control variables and extraneous variables — what you hold constant and what you cannot
The objective this slide answers, one week late
The week before this one states an objective of evaluating types of variables. The delivery audit for that week records the objective as wholly undelivered, and notes that the word "variables" occurs exactly once in the entirety of that week's material.
This slide is the nearest thing in the supplied material to that objective — two types, two synonyms each — and it appears a week later, under a slide title about defining variables, in a week whose own stated objective is about evaluating methods of data collection and does not mention variables at all. Neither week cross-references the other.
Getting it right before you collect
Four situations and what to do in each
A definition, and the distinction that decides what you may claim
The hedged treatment on this page has since been joined by two things it lacked. A later week defines a variable directly — "characteristics of an item or individual; what you analyse when you use a statistical method" — and its set reading supplies a variable-selection procedure and the association-versus-causation rule.
What to carry forward
- Dependent is what you are trying to influence or predict; independent is what you use to do it. The alternative names — response and explanatory — are often clearer.
- The source supplies a hint for telling them apart, hedged three times, and no definition of either term or of "variable" itself.
- The roles follow from your research question, not from the data, which makes the direction of the relation a claim you have to be able to defend.
- Eight further variable types in ordinary use are absent from this material entirely.
- Nothing connects the two types to the three levels of measurement given later, although the following week's coding instruction assumes the connection.
- The material sits a week after the objective it answers, and neither week refers to the other.
Frequently asked questions
How do I tell which of my variables is dependent?
Ask whether you are trying to influence or predict it from something else. If you are, the source says it is probably the dependent variable, also called the response variable. The variable you are using to do the predicting is the independent, or explanatory, variable.
Does the supplied teaching source define these terms?
No. It offers what it calls a useful hint for determining which variable is which, hedged with "probably", "typically" and "sometimes called". It does not define independent variable, dependent variable, or the word variable itself anywhere in this week or the next. Take a definition from a cited methods text if your chapter needs one.
Can the same measure be independent in one study and dependent in another?
Yes, and the source's own wording allows it — the hint asks which variable is which "in a study", and every clause turns on what the researcher is trying to do. That is why the assignment is an assertion about how you think the situation works and needs a justification from your literature.
What about confounding, moderating or control variables?
None of them appears anywhere in the supplied material. Neither do categorical, continuous, discrete or extraneous variables. The material delivers two types, and treating that as the full set is the most likely way to get caught out at examination.
My project is qualitative. Do I still need to define variables?
The week's five-step collection sequence opens by telling you to define the variables, and the same week's qualitative approaches produce themes rather than variables. No alternative first step is supplied. State in your methodology what you intend to reach conclusions about in the vocabulary your approach uses, and note that the step does not transfer.
How do variable types relate to levels of measurement?
The supplied material never says. It names two variable types on one slide and three levels of measurement ten slides later, and connects them nowhere — even though the following week instructs that coding is largely dictated by how a variable has been measured. That connection has to come from a second source.
References and source attribution
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
- Veal, A. J. 2005, Business Research Methods: A Managerial Approach, Longman.
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
- Levine, D., Berenson, M., Krehbiel, T. & Stephan, D. 2011, Statistics for Managers, 6th ed., Global Edition, Pearson Higher Education — cited in the supplied source for the five steps in data collection, the first of which is to define the variables.
- The supplied teaching source: the week 10 slide deck on collecting data, whose variables slide is reproduced in full on this page, together with both copies of the week 10 study notes.
Suggested questions for Ask KEVOS
- Identify the dependent and independent variables in my research question.
- Help me justify the direction of the relationship I am proposing.
- Which of my draft survey questions measure a variable I have actually named?
- What variable types does this teaching material leave out, and where should I read about them?
- Rewrite my research question so the dependent variable is unambiguous.