Research methods in Psychology
Although you might not think about it immediately, how you conduct psychological research is at least as important (in fact more important) than subject of the research you conduct. Here is a quick introduction to some of the key ideas in psychological research.
Formulating research questions - aims and hypotheses
Aim – a general statement about the purpose of an investigation
Experimental Hypothesis – a precise, testable statement about the expected outcome of the experiment. A hypothesis must be:
Null Hypothesis - written alongside the main hypothesis in order to make the scientific prediction complete. A null hypothesis predicts that any differences or similarities between the sets of results in an experiment are due to chance alone. An example of a null hypothesis: There will be no difference in the reaction time taken to press a button upon seeing a green square on the computer screen (measured in milliseconds) before consumption of three units of alcohol and after consumption of three units of alcohol. Any difference in results is due to chance alone. Note that the variables are clearly operationalised (it is clear how we would measure them), a prediction is made and it could be easily tested. |
|
|
Extraneous and confounding variablesAn extraneous variable is a variable other than the IV that might affect the DV if it is not controlled.
Extraneous variables must be carefully and systematically controlled so they don’t vary across any of the experimental conditions or, indeed, between participants. When designing an experiment, researchers should consider three main areas where extraneous variables may arise:-
If extraneous variables are not controlled, then can become... confounding variables. |
Confounding Variables - variables that do actually have an effect on the DV. A confounding variable could be an extraneous variable that has not been controlled.
If the IV is the only thing that is changed then it must be responsible for any change in the DV.
(N.B. - if we fail to identify & control for an extraneous variable, and we only notice afterwards that it has affected our results, then it becomes known as a confounding variable.
EXAMPLE: if researchers wished to investigate the effect of background music (condition 1) or silence (condition2) on homework performance using two classes, they’d have to control a number of possible extraneous variables. These might include age, homework difficulty and so on. If these were all successfully controlled, then the results would probably be worthwhile. However, if the researchers discovered that those in condition 1 were considerably brighter than those in condition 2, and then intelligence would be acting as a confounding variable. The researcher could no longer be sure whether any differences in homework performance were due to the presence of the music or due to intelligence levels. Results would be confounded and worthless.
If the IV is the only thing that is changed then it must be responsible for any change in the DV.
(N.B. - if we fail to identify & control for an extraneous variable, and we only notice afterwards that it has affected our results, then it becomes known as a confounding variable.
EXAMPLE: if researchers wished to investigate the effect of background music (condition 1) or silence (condition2) on homework performance using two classes, they’d have to control a number of possible extraneous variables. These might include age, homework difficulty and so on. If these were all successfully controlled, then the results would probably be worthwhile. However, if the researchers discovered that those in condition 1 were considerably brighter than those in condition 2, and then intelligence would be acting as a confounding variable. The researcher could no longer be sure whether any differences in homework performance were due to the presence of the music or due to intelligence levels. Results would be confounded and worthless.
Populations and sampling
Once we have our aim and hypothesis, we have to decide who we want to do our research on. Research will often only be relevant to certain groups of people (all females, or all teenagers, or people suffering from depression etc). We call the group of people we want to apply our research to the target population.
Sampling methods and their evaluations...
Opportunity sampling
Is not really a true method of sampling because it means taking whoever is available. Researchers take whoever they can find to take part. The way participants are selected is not systematic or structured. Psychology students tend to use opportunity sampling as they have limited access to participants Strengths
Weaknesses
|
Random sampling
Here every member of the target population has an equal chance of being selected. Everyone in the target population is available for selection each time a participant is picked out. It could be done by drawing straws or pulling names from a hat. One popular method is to give each member of the target population a number and then to take numbers from a random number table. Strengths
Weaknesses
|
Stratified sampling
Stratified sampling involves classifying the population into categories and then choosing a sample which consists of participants from each category in the same proportions as they are in the population. For example, if you wanted to carry out a stratified sample of students from a sixth form college you might decide that important variables are sex, 1st or 2nd years, age, have a part-time job and so on. You could then identify how many participants there are in each of these categories and choose the same proportion of participants in these categories for your study. Strengths
Weaknesses
|
Systematic Sampling
Made up of participants chosen mathematically. This is done by taking every nth person in the sampling frame Strengths
Weakness
|
Types of experiment in Psychology
Laboratory ExperimentsThese take place in either a lab or in a controlled environment setting, which is unnatural for the participants. They attempt to control all variables except the IV.
By changing one variable (the IV) while measuring another (the DV) while we control all others, as far as possible, then the experimental method allows us to draw conclusions with far more certainty than any non-experimental method. If the IV is the only thing that is changed then a cause and effect relationship can be found between the IV ad the DV. |
Field ExperimentsSometimes it is possible to carry out experiments in a more natural setting, i.e. in ‘the field ’. A famous example of this is the series of studies carried out by Piliavin et al (1969) in which they arranged for a person to collapse on an underground train and waited to see how long it was before the person was helped.
As with the laboratory experiment, the independent variable is still deliberately manipulated by the researcher. However it is not possible to have such tight control over variables in the field, but it does have the advantage of being far less artificial than the laboratory. |
Quasi ExperimentsMay take place in the lab or field. Like other experiments they have an IV but in this type of experiment the experimenter does not directly manipulate the IV.
Some IVs are not open to manipulation as some conditions are pre-decided by fixed characteristics. E.g. comparing men and women’s driving skills, they cannot be randomly allocated to be male or female. The IV is naturally occurring. Other examples of pre-existing variables might be age, IQ, position in the family and social background. |
Think like a Psychologist - evaluating types of experiment
Evaluating the different types of experiment is very straightforward, as soon as you understand that THE ADVANTAGES OF ONE DESIGN ARE THE DISADVANTAGES OF THE OTHER. If you understand this then you actually don't have to revise as much information, as you can reuse the same ideas in a number of different places!
For example, a lab experiment will try to tightly control variables in an artificial setting. This means that it is less likely that any confounding variables will affect the results (so cause and effect relationships can be discovered between the IV and DV), but is also means that the situation will have very low ecological validity. This means that the situation does not resemble real life, so there is a danger that the behaviour produced by participants will not be realistic as a result. If this is the case, then we cannot generalise the results of the experiment to real life behaviour (in effect the results of our experiment are worthless as they don't tell us anything about the real world). A field experiment will be the opposite. It will have high ecological validity as it occurs in a natural setting, but the poor control of extraneous variables makes it much more likely that they could affect the DV and confound the results. A quasi experiment has the obvious drawback that the researcher is not properly in control of the variables involved, so again there is the possibility for the results to be confounded. |
Want another example of a place where you can save yourself work by just understanding that the strengths and weaknesses of different ideas will be related...? Look at the evaluations of the experimental designs section below, or of the sampling methods above! The same patterns occur...
Experimental designs
As well as the type of experiment that you conduct, there are a number of ways that experiments can be designed. What we mean by this is the way that the experiment puts people into groups. The three main examples of this are below:
Non-experimental methods
Experiments are not the only way that psychologists can investigate behaviour. There are many other non-experimental methods which are widely used in the subject, such as correlation studies, content analyses, observational studies, case studies, questionnaires and interviews. Bandura et al (1961), Piliavin et al (1969) and Rosenhan (1973) are all good examples of observational studies from the CIE course.
Observational studiesOne of the simplest research methods, this simply involves observing and recording the behaviour that occurs. However, in order to make the process more scientific, a number of checks are often put in place...
INTER-OBSERVER RELIABILITY: the extent to which there is agreement between two or more observers involved in observations of behaviour. A good study should have at least 80% agreement between observers. For a recap on the meaning of the term 'reliability' in Psychology, see here. Observation studies can be participant observations, where the researcher joins the group being studied, or non-participant observations, where the researcher stays apart and observes from a distance. |
CorrelationsSometimes psychologists are interested in whether there is a relationship between two factors or variables, e.g. is there a relationship between how extrovert you are and how good at maths you are. In a case like this we might use a correlation.
In a correlation study the experimenter does not make any attempt to manipulate variables (so there is no IV or DV), he simply measures two things (e.g. maths scores and extroversion) and then compares them for a relationship (e.g. does it seem to be that as maths scores increase, so do extroversion scores). The difference between an experiment and a correlation: Experiment
Correlation
|
Case studiesA case study involves a detailed investigation of a single individual or small group of individuals. Example of the type of research that would lend itself to a case study are investigations into the effects of a stroke on later personality and behaviour, studying the effects of severe deprivation and the possibilities for recovery and so on.
Case studies often involve the use of interviews with the individual and family, friends, medical professionals etc. They may continue for many years and for this reason are often expensive and time consuming. However, they may lead to the way to future research, as they will collect extremely in-depth data. Freud's case study of Little Hans and Thigpen and Cleckley's report of 'The three faces of Eve' are both examples from the CIE specification of a case study. |
Test your understanding...With this quiz on research terms... or this one...
... and with these Holah quizzes 1. Matching quiz on variables. 2. This mini-test checks out your knowledge of IVs and DVs 3. Here is a match the type of experiment quiz. 4. Here is a match the design quiz and here is another one. 5. Here is a cloze exercise on hypotheses for experiments. |
Applying your understanding to a real study - The Marshmallow Test
|
First watch the video to the left, then open the document below. In
the summary of the experiment table of the document, fill in the gaps in the description of the
study with the appropriate answers, based on what we’ve done in the last few
lessons.
TIPS!
When evaluating an experiment, think of the positives and negatives of the ways that the experiment is designed. For example, what are the pros and cons of the sampling method used? What about the experimental design? All of these problems will affect the study as well. |
Want more practice...? Firefly is your place
Log into the JIS Firefly page and find the 'Research methods' section in the 'Psychology' pages ('Humanities' section) for lots more resources and practice opportunities.