INTRODUCING VARIABLES
A variable is something that varies – a factor that changes. Variables can be things like a person’s race or gender, the type of weather going on at the moment or the time of day, or what sort of exam results different people get or their scores on tests.
INDEPENDENT VARIABLES
In an experiment, a scientist will focus on a particular thing they’re interested in – the independent variable (IV). This is the topic the scientist is exploring, the thing that might be the hidden cause of behaviour.
You might like to remember that scientists are INterested in the INfluence of the INdependent variable Manipulated IVs are things that the scientists can change in order to see what happens. For example, a scientist could change how noisy a room is to see if people worked harder or not. Another example of a manipulated variable might involve testing a drug on one group of people while leaving another group normal. A naturally-occurring IV is something that changes of its own accord. For example, no scientist can “make” people male or female. Sometimes it’s physically impossible to manipulate a naturally-occurring variable. Sometimes it’s just unethical – for example, it would be wrong to give someone a nasty disease just to see how they coped with it.
DEPENDENT VARIABLES
In an experiment, a scientist will be looking for a certain type of result. The dependent variable (DV) is the thing that changes – the outcome that you measure. In a successful experiment, changes in the IV should cause changes in the DV.
Think of it this way: the dependent variabe depends on the IV Remember how science tries to discover laws and make predictions? That’s the beauty of experiments; they show causation - cause-and-effect. If your DV changes at the end of an experiment, you can conclude that the IV was causing it to change. For example:
OPERATIONALISING YOUR VARIABLES
Sloppy or vague research looks at variables like "memory" or "intelligence" and compares cariables like "age" or "role-models".
Operationalising means stating your IV and DV in ways that make it obvious how they are being manipulated or measured. In particular, an operationalised DV will create quantitative data. For example:
EXTRANEOUS & CONFOUNDING VARIABLES
Of course, you can’t always tell cause-and-effect from an experiment, especially an experiment on people. This is because the IV isn’t the only thing having an effect on the DV. Usually there are lots of other factors going on too.
Factors other than the IV and the DV that might change during the course of a study are the extraneous variables. Extraneous variables fall into three types:
Most extraneous variables make no difference or even cancel each other out, but they "sow seeds of doubt" about whether the DV is really being caused by the IV - they lower the internal validity of the study. Some extraneous variables definitely make a difference to the behaviour of participants and interfere with the DV. These are called confounding variables because they confound (spoil) the validity of the research. CONTROLLING EXTRANEOUS VARIABLES
To deal with the problem of confounding variables, scientists will use experimental controls. A control is something that stops a confounding variable from interfering with the DV because it makes sure that the only important variables changing are the IV and the DV.
Here are some examples:
It's much easier to set up controls for extraneous variables in a lab experiment, because everything is taking place in a controlled setting that you can arrange beforehand.
In a field experiment, it's harder to set up controls because you have less control over a real-life setting with ordinary events going on all round you. Although controls increase the internal validity of a study by reducing confounding variables, they can have the effect of lowering the external validity just as much. In particular, highly controlled experiments can lack ecological validity if the setting or the task become too unfamiliar or artificial. This is a trade-off between internal and external validity that all psychologists must make. DEMAND CHARACTERISTICS
One type of extraneous variable that gets special mention is demand characteristics. This is when participants behave unnaturally because they believe they know the purpose of the research they are taking part in.
Martin Orne (1962) coined this phrase when he pointed out how people are naturally curious and in an experiment they tend to wonder what it’s all about. If people think they’ve figured out what the purpose of the experiment is, the temptation is to be “good participants” and make the experiment successful. Other people are awkward customers and, when they think they know what the experiment’s about, they try to ruin it. This is what Phil Banyard (2001) calls the “Screw You” Effect - another type of demand chaacteristics. Demand characteristics lower the internal validity (because something other than the IV is causing the DV) and external validity (because these people wouldn't behave that way in real life) as well as the predictive validity (because you can't predict future behaviour from unnatural behaviour). Naïve participants do not realise they are in a study - or if they do know they are in a study, they have been misled about what it's really all about. Misdirection is an experimental control that reduces demand characteristics but it raised ethical problems because the naïve participants cannot give informed consent if they've been misled about the point of the study.
|
CONTROLLING VARIABLES IN PSYCHOLOGY
|
EXEMPLAR ESSAY
|