Methods of Research in Psychology: Lab experiments, Observations, and Questionnaires
Edited by Jamie (ScienceAid Editor), SpellBot, Taylor (ScienceAid Editor)
In Psychology, this is one of many methods we use to investigate something. The key things that make it an experiment are that it is usually carried out in a controlled environment which will probably be a laboratory; where key variables are manipulated. The experimental approach is the sort of approach that is used almost exclusively in other sciences like Chemistry.
Variables are a key feature and there are two types that we have. The independent variable is the condition that is manipulated. For example, in the Levels of Processing experiment, this was the depth of processing.
And then there is the dependent variable = that is what you're measuring. In the levels of processing experiment this was recall of words. In the experiment everything except the independent variable is kept the same; but factors you haven't controlled that could affect your results are called extraneous variables.
A further feature of lab experiments are experimental designs, which outlines how you use your participants. The table below goes through some of the different experimental / research designs.
|Repeated Measures||The same group of people are used in both the conditions.|| Same people are used so there are no differences in characteristics between the two. So there are no participant variables.
Uses fewer participants which reduce cost and makes it easier to carry out the experiment.
| There are practice effects where participants are already aware of how to do the task. Also order effects where the order that you do something in could affect the outcome.
These two issues can remedied by counterbalancing where half the participants do A then B and the other half do B then A.
|Independent Samples||Different participants are used in each of the conditions.|| There are no order or practice effects.
Exactly the same materials can be used.
| The participant variables are not controlled at all.
Using two groups of people mean that more resources have to be used.
|Matched Pairs||Two separate groups of people are used, but they are matched up so that on some variables (like sex, age, class) there is someone exactly the same on the other side.|| No order or practice effects.
Characteristics are controlled.
| Complicated to match people up.
Not possible to have two people exactly the same so two groups can't be said to be exactly the same.
An observation is where a Psychologist will go into a situation and look at how people behave. There are two types: participant and non-participant observations.
In a non-participant observation the researcher sits outside of the situation and watches. It can either be disclosed where the people being observed know they're being watched or undisclosed. The advantages of this method are that it looks at behaviour in a natural setting so people act more realistically. On the downside, there is no control of how people act, so it may not actually reflect how they feel, and informed consent is difficult to get.
Now we come to participant observation this is where the researcher is involved in the situation they are observing; for example, going 'under cover'. The advantages of this method are that normal behaviour is seen, and is often the only way of finding out about certain things (like criminal activity for example). Disadvantages are that the researcher may influence what happens, it is only a one off situation and not completely representative, and it uses a lot of resources.
We are all familiar with questionnaires; but it is a lot more difficult than you think. Again, there are two types of questions:
- 1Closed.A question is asked and several options are given. This is good, because it makes questionnaires easier to analyze, but doesn't give details.Advertisement
- 2Open.The questions asked require a written response in the space provided.Advertisement
Designing questions for a questionnaire is more difficult than it first seems, and there are several things that you have to be careful to avoid. Some of these are outlined in the table below.
|Leading Questions||This means wording the question so the answerer is manipulated to have the participant respond in a particular way. For example: Most people believe global warming is caused by carbon emissions; do you agree?|
|Ambiguity||Questions that could be interpreted in different ways, like: What is your mood?|
|Complicated||This might involve be giving a lot of information before the question is asked, phrasing it in a difficult way; or including jargon like scientific terminology that a lot of people won't understand.|
|Lots of Questions||You may want to learn a lot from the questionnaire, but if you have too many questions, the answerer will get fed up and won't respond properly, usually by giving neutral responses.|
- If you have problems with any of these steps, ask a question for more help, or post in the comments section below.
Categories : Approaches
Recent edits by: SpellBot, Jamie (ScienceAid Editor)