Research Methods

By the end of this section you will be able to describe a range of research methods:

• Observation
• Case-study
• Survey
• Correlation longitudinal and cross-sectional studies
• The advantages and disadvantages of these methods.

Experiment

The relationship between two things is explored deliberately producing a change in one variable (the independent variable IV) and recording what effect this has on the other variable (the dependent variable DV).

Strengths

• Determines a causal link
• Can be well controlled
• Can be replicated

Weaknesses

• Artificial and therefore may not generalise to real-life,
• Total control is never possible, many variables remain beyond the control of the experimenter
• Can be affected by, for example, experimenter bias, demand characteristics, volunteer bias, sample bias.
• Some classes of participants, such as children, react poorly under experimental conditions.
• In some cases it would be unethical to manipulate exposure to certain conditions, e.g. early deprivation in children.

Example of an experiment:

Field experiment

An experiment conducted in the natural environment, where the participants are unaware that they are participating in a psychology experiment. The independent variable is still manipulated.

Uses:

Where the research design calls for a natural setting (e.g. library, street, bus).

Strengths

• Greater validity
• Avoids participant bias and demand characteristics

Weaknesses

• Extraneous variables are harder to control
• More time-consuming and expensive

Natural experiment

This is used where conditions may vary naturally, the effects of an independent variable can be noted without any intervention by the experimenter. It is still an experiment in the sense that a cause and effect are being identified. If a new teaching programme is being tested, a comparison might be made before and after.

Strengths

• Greater validity
• Avoids participant bias and demand characteristics

Weaknesses

• Participants may be aware of being studied (possible Hawthorne effect)
• Loss of control over extraneous variables
• Since we have not directly manipulated the independent variable we cannot truly claim cause and effect; a third variable may be involved.

Experimental design

Repeated methods

The same participant is tested in both conditions of the experiment.

Strengths

• No individual differences
• Uses fewer participants
• Less time consuming and expensive

Weaknesses

• Order effects
• Participant may guess the purpose of the experiment

Independent groups

The participants are put into groups each group takes part in only one condition of the experiment.

Strengths

• Avoids order effects
• Participants will not guess the purpose of the experiment

Weaknesses

• Needs more participants
• Introduces individual differences
• More time consuming and expensive than repeated measures.

Matched pairs

The participants are matched, by testing before the experiment on key attributes. They are then randomly assigned to the two conditions of the experiment.

Strengths

• No order effects
• Individual differences controlled

Weaknesses

• Matching is difficult
• Needs a lot of participants
• It is time consuming and expensive

Non-experimental methods

Case Studies

This method involves in-depth and detailed study of an individual or particular group. The case study method is often applied to unusual or valuable examples of behaviour that may provide important insights into psychological function or refutation of psychological theory. Example; Clive Waring in the study of memory.

Strengths

• Highly detailed and in depth data is provided that superficial methods might miss or ignore.
• High ecological validity of data obtained.
• Often the only method suitable for studying some forms of behaviour e.g. investigating the acquisition of human language in primates.
• Often the only method possible due to rarity of behaviour, e.g. natural cases of human environmental deprivation, such as the case of Genie.

Weaknesses

• No cause and effect can legitimately be inferred.
• Lack of generalisability to the population due to single cases being too small and unrepresentative a sample.
• Low reliability due to:
  • Many case studies involving recall of past events, which may be open to memory distortion
  • Participant reactivity
  • Lack of observer objectivity
• Difficult or impossible to replicate
• Time consuming and expensive

Correlations

A correlation is not a research method. This is a method of data analysis that measures the relationship between two or more variables to see if a trend or systematic pattern exists between them. Examples of correlations aggression scores and TV viewing from a sample of children or attractiveness ratings of males and females from a sample of newly wed couples.

Strengths

• Precise information on the degree of relationship between variables is available in the form of the correlation coefficient. It can readily quantify observational data.
• No manipulation of behaviour is required.
• Strong significant correlations can suggest ideas for experimental studies to determine cause and effect relationships.

Weaknesses

• No cause and effect can be inferred
• Correlations should be plotted out on scattergrams to properly illustrate the relationship between variables

Observations

These involve watching and recording the behaviour of interest. The observer may make written records or use video or audio-recording for more detailed analysis later. It is a good idea to make more than one observation of the same thing. The two sets of observations can then be checked to see if they agree or are consistent. If they do, this is known as reliability and it helps us to have confidence in the results.

Naturalistic Observation

Naturalistic observations involve the recording of spontaneously occurring behaviour in the participants own natural environment.

Strengths

• High ecological validity of observed behaviour if observer is hidden
• Can be used to generate ideas for or validate findings from experimental studies
• Sometimes the only ethical or practical method.

Weaknesses

• Cannot infer cause and effect relationships between variables that are only observed but not manipulated.
• Lack of control over conditions makes replication more difficult.
• Ethical problems of invasion of privacy.

Controlled Observation

Controlled observation involves the recording of spontaneously occurring behaviour, but under conditions contrived by the researcher (e.g. in the laboratory).

Strengths

• More control over environment which leads to more accurate observations.
• Greater control leads to easier replication.
• Usually avoids ethical problems of consent, unless research purpose and observer are hidden.

Weaknesses

• Participant reactivity may distort the data if the participant is aware of being observed e.g. abnormal sleep patterns in unnatural laboratory conditions.
• Lower ecological validity than naturalistic observations, can cause demand characteristics.
• Cause and effect cannot be inferred.

Participant Observation

Participant observations involve the researcher becoming involved in the everyday life of the e.g. group members, either with or without their knowledge.

Strengths

• Very high ecological validity if observer undisclosed, less if disclosed depending upon level of integration with participants.
• Extremely detailed and in depth knowledge available, not gained from any other method.

Weaknesses

• Difficult to record data promptly and objectively, and impossible to replicate exactly.
• Observers behaviour may influence group members.
• Cause and effect cannot be inferred.

Interviews

Structured interviews

These contain fixed predetermined questions and ways of replying (e.g. yes/no)

Strengths

• Easy to quantify and analyse.
• Reliable, replicable and generalisable.

Weaknesses

• Less validity
• distorts/ignores data due to restricted answers or insensitivity.

Semi-structured interviews

These contain guidelines for questions to be asked, but phrasing and timing are left up to the interviewer and answers may be open-ended.

Strengths

• Fairly flexible, and sensitive.
• Fairly reliable and easy to analyse.

Weaknesses

• Flexibility of phrasing and timing could lead to lower reliability.
• Open-ended answers are more tricky to analyse.

Clinical interview

These are semi-structured guidelines but further questioning to elaborate upon answers.

Strengths

• Very flexible, sensitive and valid
• Fairly reliable and easy to analyse

Weaknesses

• Flexibility leads to more difficulty in replication and bias from interviewer.

Unstructured interview.

These may contain a topic area for discussion but no fixed questions or ways of answering. Interviewer helps and clarifies interview.

Strengths

• Highly detailed and valid data.
• Extremely flexible, natural and unconstrained.

Weaknesses

• Very unstandardised, therefore not replicable, reliable or generalisable.
• Difficult to quantify and analyse.

Questionnaires

Opinion surveys

Questionnaires are written methods of gaining data from subjects that do not necessarily require the presence of a researcher. They include:

Opinion surveys e.g. attitudes scales and opinion polls. Questions can be closed or open-ended and should be precise, understandable and easy to answer.

Strengths

Highly replicable and easy to score (unless open-ended answers).

Weaknesses

Biased by socially desirable answers, acquiescence 9agreeing with items) and response set (replying in the same way).

Longitudinal and cross-sectional methods

If a psychologist is interested in the ways in which behaviour changes over time these are the two main research strategies available.

Longitudinal studies

This involves following up the same group of individuals on a number of different occasions over time. This could be daily, weekly, monthly, annually. The television programme 35 Up has revisited a group of boys and girls from widely differing social backgrounds at 7, 14, 21, 28 and 35.

Strengths

• The psychologist can study age-related changes in behaviour
• The impact of early experience on later behaviour
• Using the same people over and over cuts down on the number of individual differences which could affect the behaviour
• Individuals can be studied in depth and detail

Weaknesses

• It can be very time consuming and expensive
• There may be a high drop-out rate of participants, possibly meaning that the sample is no lon
ger representative
• Research interests change over time so a study may literally be out of date before it is over

Cross-sectional studies

This method involves taking samples of individuals to represent different age groups and studying them at the same time.

Strengths

• It is usually quicker and cheaper than the longitudinal method
• It helps us to see age related changes in behaviour

Weaknesses

• Individual differences between age groups which could obscure the results
• The information gathered may be more superficial than the longitudinal method
• As participants are seen just once, their behaviour at that time may not be typical
• It may exaggerate differences between age groups and give a false impression of stops and starts in development rather than a smooth progression