Research Methods

When designing a questionnaire, there are several ways you can approach the study:

Use closed questions (fixed choice of answers), to generate data for easy analysis.

Use open questions (space to write any answer) for more detailed individual answers.

Keep questions and instructions clear and easy to understand.

Ask purposeful questions to help find information needed for the study.

Pre-code closed questions for quick analysis of the answers.

Carry out a pilot study first, a test run, making changes if needed.

Use attitude scales to test strength of feeling.

Strengths:	Weaknesses:
Many people can be tested quickly. It is easy to generate quantitative data and easy to analyse.	Social desirability - people say what they think looks good.
Used to collect large amounts of data about what people think as well as what they do!	People may not tell the truth, especially on sensitive issues, for example, sexual behaviour.
Convenient - researcher does not need to be present as answers can be mailed so respondent has time to consider answers.	If researcher is present then this may affect answers. Also, postal surveys may have low response rate.
Can quickly show changes in attitudes or behaviour before and after specific events.	Difficult to phrase questions clearly, you may obtain different interpretations of questions.

Interviews are face-to-face conversations, these can be unstructured, apparently informal chats, or they can be formal, structured interviews with pre-determined questions. For example, clinical tests used in psychiatry.

Interviews are recorded for later, in-depth analysis.

Strengths:	Weaknesses:
Detailed information can be obtained and avoids oversimplifying complex issues.	Difficult to analyse if unstructured and qualitative in nature.
Greater attention to individual's point of view this is important in clinical psychology.	Time-consuming, expensive.
Unstructured, casual interviews may encourage openness in answers.	Possible interviewer effects. For example, people affected by attractiveness of interviewer!

Research can be described as quantitative or qualitative.

Quantitative research: Gathers data in numerical form and is concerned with making 'scientific' measurements. Quantitative data analysis uses a barrage of inferential statistical tests.

Qualitative research: Gathers information that is not in numerical form. For example, diary accounts, open-ended questionnaires, unstructured interviews and unstructured observations.

Qualitative research is useful for studies at the individual level, and to find out, in depth, the ways in which people think or feel.

Analysis of qualitative data is difficult and requires accurate description of participant responses, for example, sorting responses to open questions and interviews into broad themes.

Quotations from diaries or interviews might be used to illustrate points of analysis.

Expert knowledge of an area is necessary to try to interpret qualitative data and great care must be taken when doing so, for example, if looking for symptoms of mental illness.

Accurate descriptions of individual behaviour patterns might be crucial to diagnosis, treatment and follow-up of a person with a mental disorder.

Below is a list of terms that are commonly used, it is important to know how to use them:

Arithmetic mean: All values in a set of data are added together and divided by the number of values (N). Used with normal distribution and interval level data. Not suitable for use where extreme values can distort the mean. The most sensitive measure of central tendency.

Median: All values are arranged in order, the middle value is the median. Used with interval or ordinal level data, the median is not affected much by extreme values.

Mode: The most frequent value or score in a set of data. Used with nominal data. Does not give any information about other values.

Range: Simple measure of dispersion- shows the total spread of data. Difference between highest and lowest scores in a set of data: top value minus bottom value plus 1. Affected by atypical, extreme values.

Standard Deviation: Measure of dispersion- shows degree of clustering of values around the mean. Calculating standard deviation (S): Square root of sum of all squared deviations from the mean, divided by N (or sometimes N-1). The most accurate measure of dispersion.

Graphs and charts give a quick visual impression of any patterns or trends in your results. They should be used to help summarise your results.

You need to remember the difference between a bar chart, a histogram and a frequency polygon.

Bar charts:

Used for nominal data (data in categories).

The x-axis (frequencies are usually on the y-axis) does not need to show a complete scale (if showing categories).

There should be gaps between the bars.

Histograms:

Used for interval or ordinal data.

No intervals (if data is grouped) are missed, even if they are empty. Class intervals are represented by their mid-point at the centre of each column.

There are no gaps between columns.

Frequency polygons:

Used for interval or ordinal data.

All class intervals are represented.

Instead of columns, a line is used to join the mid-point of each class interval.

See how you get on with the exercise below. Drag the three pink boxes onto the correct blue box and then mark your answer:

OK, so you've thought up this brilliant psychological experiment and designed it perfectly.

But who are you going to try it out on?

Sampling techniques are very important. Several different types of sampling are used according to the type of study and the subjects you want to target.

Random sampling: Everyone in the entire target population has an equal chance of being selected.

Opportunity sampling: Uses people from target population available at the time.

Systematic sampling: Chooses subjects in a systematic way. For example, every 10^th person from a list or register.

Self-selected sample: Participants volunteer. For example, by answering an advert.

Stratified sampling: Divides target population into groups, people in sample from each group in same proportions as population. So you would have a higher number of people between the ages of 20-30 than 70-80.

This does not mean becoming emotionally involved and going out with participants!

What it does mean is ways in which either the researcher or the participants can influence the results this is known as "causing bias".

Researcher effects: Researcher can affect the behaviour of the participants, thus affecting the results of the study.

For example:

The researcher might unwittingly communicate his expectations to the participants. This could happen through only small changes in body language or tone of voice.

Or it can be in the interpretation of data, a researcher may read into things more of what he or she would like to find!

An attractive researcher might affect participant responses. For example, male researchers smile at female participants more than they do at male ones!

Even rats learned mazes faster when expected to! (Rosenthal, 1966)

Just the presence of the researcher can affect participant behaviour, more so if the researcher is filming people.

Demand characteristics: Participants might read things into the situation and start changing their behaviour they respond to the perceived demands of the study.

Participants may worry about being in a psychological study and want to appear 'normal', this may change their behaviour.

Participants may try to guess what the investigation is about then behave in the way they think the investigator wants them to.

On the other hand, they may deliberately try to behave in an unexpected way. Unofficially known as the "f*** you effect").

Participants might just try to 'look good' (social desirability) and behave out of character or not tell the truth. This can be a problem for questionnaires on sensitive issues.

There are several things a researcher can do:

Disguise the purpose of the investigation: There is some deception in many psychological studies to stop participants guessing the aims and changing their behaviour. Participants in Milgram's obedience studies thought it was a study on effects of punishment on learning and memory.

Single-blind design: Participants do not know which condition (experimental or control) they are in. For example, the use of placebos in trials of drug treatments.

Double-blind design: Neither the participants nor the experimenter know which condition people are being treated to. For example, a research assistant giving out drugs and measuring their effects does not know who has the placebo and who has the drug.

Independent variable (IV): Variable the experimenter manipulates - assumed to have a direct effect on the dependent variable.

Dependent variable (DV): Variable the experimenter measures, after making changes to the IV which are assumed to affect the DV.

Extraneous variables (Ex Vs): Other variables, apart from the IV, that might affect the DV. They might be important enough to provide alternative explanations for the effects, for example, confounding variables.

Laboratory experiment: Artificial environment with tight controls over variables.

Field experiment: Natural environment with independent variable manipulated by researchers.

Natural experiment: Natural changes in independent variable are used - it is not manipulated.

Note: In a true experiment participants are randomly allocated to groups.

Think of some examples from your course.

Which categories would the following studies fall into?

Laboratory experiments

Strengths:	Weaknesses:
Tighter control of variables. Easier to comment on cause and effect.	Demand characteristics - participants aware of experiment, may change behaviour.
Relatively easy to replicate.	Artificial environment - low realism.
Enable use of complex equipment.	May have low ecological validity - difficult to generalise to other situations.
Often cheaper and less time-consuming than other methods.	Experimenter effects - bias when experimenter's expectations affect behaviour.

A field experiment takes place anywhere in a natural setting; it could take place in a school, hospital, the street or an office.

Note:

A field experiment is an experiment; the independent variable is manipulated. Not all field studies are experiments.

Strengths:	Weaknesses:
People may behave more naturally than in laboratory - higher realism.	Often only weak control of extraneous variables - difficult to replicate.
Easier to generalise from results.	Can be time-consuming and costly.

Strengths:	Weaknesses:
Situations in which it would be ethically unacceptable to manipulate the independent variable.	The independent variable is not controlled by the experimenter.
Less chance of demand characteristics or experimenter bias interfering.	No control over the allocation of participants to groups (random in a 'true experiment').

Three experimental designs are commonly used:

Independent groups: Testing separate groups of people, each group is tested in a different condition.

Repeated measures: Testing the same group of people in different conditions, the same people are used repeatedly.

Matched pairs: Testing separate groups of people - each member of one group is same age, sex, or social background as a member of the other group.

In each case, there are one or more experimental groups, where the independent variable has changed and a control group where the independent variable has not changed.

Independent groups:

Avoids order effects. If a person is involved in several tests they man become bored, tired and fed up by the time they come to the second test, or becoming wise to the requirements of the experiment!

More people are needed than with the repeated measures design.

Differences between participants in the groups may affect results, for example; variations in age, sex or social background. These differences are known as participant variables.

Repeated measures:

Avoids the problem of participant variables.

Fewer people are needed.

Order effects are more likely to occur.

Matched pairs:

Reduces participant variables.

Avoids order effects.

Very time-consuming trying to find closely matched pairs.

Impossible to match people exactly, unless identical twins!

Note:

Counterbalancing: Alternating the order in which participants perform in different conditions of an experiment. For example, group 1 does 'A' then 'B', group 2 does 'B' then 'A' this is to eliminate order effects.

Randomisation: Material for each condition in an experiment is presented in a random order, this is also to prevent order effects.

Correlation is a statistical technique used to quantify the strength of relationship between two variables.

Used a lot in psychology investigations, for example Murstein (1972) carried out a correlation analysis of ratings of attractiveness in partners ('computer dance' study).

Strengths:	Weaknesses
Calculating the strength of a relationship between variables.	Cannot assume cause and effect, strong correlation between variables may be misleading.
Useful as a pointer for further, more detailed research.	Lack of correlation may not mean there is no relationship, it could be non-linear.

For a correlational study, the data can be plotted as points on a scattergraph. A line of best fit is then drawn through the points to show the trend of the data.

If both variables increase together, this is a positive correlation.

If one variable increases as other decreases this is a negative correlation.

If no line of best fit can be drawn, there is no correlation.

Correlation can be quantified by using a correlation coefficient - a mathematical measure of the degree of relatedness between sets of data.

Once calculated, a correlation coefficient will have a value from -1 to +1.

+1 = perfect positive correlation all points on straight line, as x increases y increases. A value close to one indicates a strong positive correlation.

0 = no correlation points show differing degrees of correlation.

Note: A correlation around zero may disguise a non-linear relationship.

-1 = perfect negative correlation all points on straight line, as x increases y decreases. A value close to -1 indicates a strong negative relationship.

Note: In real life human situations, or psychology experiments you will not find perfect correlation between variables, life is just like that.

What psychologists do is calculate a correlation coefficient, then, using statistical tables (thought up by brilliant mathematicians) work out the probability that their results could have occurred at random.

If they can say there is a 95% chance of their results really showing a strong correlation, then they accept that there is one.

Watching the behaviour of humans or animals in a natural environment.

The researcher does not manipulate variables and does not interfere with things - they try to remain inconspicuous.

Examples:

1. Rosenhan (1973): Pseudopatients admitted to psychiatric hospitals and treatment by hospital staff observed.
2. Lorenz (1937): Famous studies on imprinting in animals.

In participant observation the observer acts as part of the group being watched.

You will need to be systematic, observations may be either structured or unstructured.

Structured observation: Uses tables of pre-determined categories of behaviour and systematic sampling.

Ways of sampling in structured observational studies:

Time sampling: Observations may be made at regular time intervals and coded.

Event sampling: Keep a tally chart of each time a type of behaviour occurs.

Point sampling: Focus on one individual at a time for set period of time.

Unstructured observations: Record everything that happens. It may be difficult to avoid bias by focusing on what you want or expect to see happening, in theory all observations are noted as anything could prove to be important.

May use a diary method to record events, feelings, or moods.

Video recording: This is useful as behaviour may be analysed in more detail later.

Strengths:	Weaknesses:
More natural behaviour occurs if people are unaware of observation.	Observer may affect behaviour if detected.
Studying of animals that cannot be observed in captivity.	Difficult to replicate - cannot control extraneous variables.
Study of situations that cannot be artificially set up.	Need for more than one observer.

You need to know about several different research techniques, when they are used, and the strengths and weaknesses of each.

This table outlines different methods used in research:

Research method:	Description:
Correlation	Statistical technique - measures strength of relationship between variables.
Experiment	An independent variable is manipulated while others controlled, to see effects on a dependent variable.
Interview	Used to gain in-depth information and individual views.
Naturalistic observation	Watching behaviour, as it occurs spontaneously, in a natural setting.
Questionnaire survey	A snapshot of large number of people's attitudes, opinions or behaviour.

The aim of an investigation is its general purpose.

What are you trying to achieve in the investigation?

The hypothesis is a precise, testable statement or prediction about the expected outcome of an investigation.

A 'null hypothesis' (Ho) prediction is one that states results are due to chance and are not significant in terms of supporting the idea being investigated.

For example:

There is no evidence that there is a difference between groups in the amount they remember.

A research hypothesis (H1) prediction is one that states that results are not due to chance and that they are significant in terms of supporting the idea being investigated.

For example:

There is evidence that there is a difference between groups in the amount they remember.

A one-tailed hypothesis is a directional hypothesis.

For example:

Instead of saying there will be a difference between groups in the amount they remember, you predict which group will remember most.

A two-tailed hypothesis is one in which the direction of results is not predicted.

For example:

You may predict a difference between groups, but have no idea which way the difference will fall.

Remembering the difference between one and two tailed hypotheses:

A one-tailed cat can only point its tail in one direction at a time!

A two-tailed cat can point its tail in both directions at once - it does not tell you the way things will go - it is non-directional!

The following factors are important to consider when designing an investigation:

A pilot study is a test run on a few participants this enables you to check for design faults before carrying out an investigation on a larger scale, this is a routine procedure especially used when carrying out questionnaire.

Reliability of results is very important, so if a study is replicated the findings should be similar.

Validity, does a test measure what it was designed to measure. For example, do IQ tests really measure 'intelligence'?

Internal validity, extent to which study is free of design faults, which may affect results.

Ecological validity this is a type of 'external validity'. This means the extent to which generalisation can be made from the test environment to other situations.