The three kinds of variable in an experiment
- The independent variable is the thing you change on purpose.
- The dependent variable is the thing you measure to see whether that change mattered.
- Extraneous variables are everything else that could affect what you measure.
The variable you change
Definition
Independent variable
The variable the researcher deliberately changes or manipulates, usually set up as two or more conditions to compare.
- The independent variable (IV) is the variable you deliberately set up in two or more versions, called conditions.
- Naming an IV properly means naming the conditions being compared, not just the general subject.
- Revision method is only half an answer, whereas revision method, either flashcards or re-reading notes, is the full one.
Example
- A class times how quickly people catch a dropped ruler in a silent room and in a room with music playing, so the IV is the sound in the room, with the conditions silence and music.
- A class tests spelling after a ten minute break and after no break, so the IV is whether a break was taken, with the conditions break and no break.
The variable you measure
Definition
Dependent variable
The variable the researcher measures to see whether changing the independent variable has had an effect.
- The dependent variable (DV) is the outcome you record once the conditions have been run.
- Naming a DV properly means naming the exact score, count or time you write down.
- Memory is not a dependent variable, but the number of words recalled out of twenty is.
Example
- In the ruler experiment the DV is the distance in centimetres the ruler falls before it is caught.
- In the spelling experiment the DV is the mark out of twenty on the spelling test.
Variables that get in the way
Definition
Extraneous variable
Any variable other than the independent variable that could affect the dependent variable. If it is not controlled, it offers an alternative explanation for the results.
- An extraneous variable is anything other than the IV that could push the DV up or down.
- Left alone, it hands anyone reading your results a rival explanation for them.
- If the flashcard group also happened to work in a quiet room, the quiet room explains their higher scores just as well as the flashcards do.
Situational variables
- A situational variable is a feature of the setting or of the way the study is run that could affect the DV.
- It does the most damage when the two conditions are run in different places or at different times.
Example
- Noise from the corridor during the afternoon session but not during the morning one.
- Instructions rushed through with one class and explained slowly to another.
Participant variables
- A participant variable is a feature of the people themselves that could affect the DV.
- It does the most damage when different people take part in each condition, because one group can end up stronger than the other purely by luck.
Example
- One group happening to contain more of the fastest readers in the year.
- Differences in how tired people are, or in how much they care about doing well.
Telling the two kinds apart
- If moving a participant to a different room would change it, it is situational.
- If it would follow them into that new room, it is a participant variable.
Common Mistake
- Do not answer noise when you are asked for a participant variable, because noise is in the room.
- Do not offer the independent variable itself as an extraneous variable, because it is the one difference you meant to create.
Turning this into marks
- For the IV, quote the two conditions the scenario actually describes.
- For the DV, give the unit or the score as well as the thing measured.
- The mark most often lost is on extraneous variables, where a bare noun is given without saying which kind it is and how it could change the results.
Self review
- What is the difference between the independent variable and the dependent variable?
- Name one situational variable and one participant variable in a class experiment on revision methods.
- Why does giving memory as a dependent variable lose the mark?
- How can you test whether an extraneous variable is situational or participant?
