Psychological investigations

Psychological investigations

Experiments

• Lab – Researcher manipulates one variable and measures the effect on another. All other variables are controlled as far as possible. Sometimes considered Cause (IV) and effect (DV).  E.g. Loftus & Palmer

-High control

-Not ecologically valid.

-Demand characteristics.

-Often involve deception – ethics.

• Field – Real-life situation. IV and DV still exist, but not in controlled way. E.g. Piliavin et al.

-Natural environment (ecological validity)

-Lack of control

• Quasi – Naturally occurring IV such as age or gender, cannot be experimentally manipulated. E.g. Maguire.

-Not high control.

-Difficult to establish causal relationships.

Extraneous variable - A variable other than the IV that affects the DV.

Experimental design

INDEPENDENT MEASURES

• Using different participants for each condition of an experiment.
• This means there is no problem with order effects.
• Error may result from individual differences.
• Uneconomic as twice the number of participants are required.

REPEATED MEASURES

• Testing the same individuals in each condition.
• No individual differences.
• Fewer participants required.
• Order effects – people behave differently according to which order the conditions were carried out (Pps may become bored/tired). Reduced by counterbalancing which randomises the order that different Pps carry out each condition.

MATCHED PAIRS

• Using different participants for each condition but participants are matched in key variables, such as scores in certain tests.
• Difficult to obtain matched pairs.
• Time consuming.
• Expensive.

Hypotheses

A predictive statement that can be tested.

ONE-TAILED HYPOTHESIS (DIRECTIONAL)

• Predicts the direction of results.
• E.g. Participants will perform better on memory tests at 10am than 10pm.

TWO-TAILED HYPOTHESIS (NON-DIRECTIONAL)

• Does not predict the direction of results.
• E.g. There will be a difference in performance on a memory test between participants tested at 10am and participants tested at 10pm.

ALTERNATE HYPOTHESIS

• States expected results
• All adult participants will gain higher scores on memory tests at 10am than 10pm.

NULL HYPOTHESIS

• States no difference.
• E.g. there will be no significant differences in performance on a memory test between participants tested at 10am and participants tested at 10pm.

Control

• Sometimes, the results of an experiment are due to the individual, e.g. If participants perform better on a test in the morning than the evening, it might be that participants who took the test in the morning are better at memory tests.
• Sometimes however, results mat be due to the situation, e.g. if the test was conducted in a noisy environment. Scientists reduce this as far as possible.
• To reduce demand characteristics, a single-blind technique can be used. Here, participants are not aware of the aims of the experiment.
• To reduce experimenter bias, a double blind technique can be used, both researcher and participant are unaware of the aim.

Descriptive statistics

• Descriptive statistics offer a summary of data.
• Do not allow us to make a conclusion.
• Two main ways:

-Measure of central tendency (mean, median, mode).

-Measure of dispersion (range, Standard deviation).

Observations

• Tend to be high in ecological validity
• Does not manipulate an IV.
• No demand characteristics.
• Useful as a starting point.
• Difficult to replicate.
• Do not give information on thoughts or feelings.
• Little control of extraneous variables so cannot conclude casue and effect.
• Observer bias.
• Lack of informed consent.
• Time consuming.

NON-PARTICIPANT OBSERVATION

• Researcher does not join in activity being observed.

PARTICIPANT OBSERVATION

• Observer is also a participant in the activity being studied.
• Provides more insights about behaviour.
• Observer may lose some objectivity.

UNDISCLOSED (COVERT) OBSERVATION

• Participants are not fully aware that they are being studied.
• Researchers may use one-way mirrors.

STRUCTURED OBSERVATION

• Researchers design a coding system to record and categorise participants behaviour.
• Generally provide quantitative behaviour.
• Sometimes behaviours may overlap or a behaviour may occur that has not been previously thought of.

UNSTRUCTURED OBSERVATION

• Researchers record behaviour they can see.
• Can be difficult without a camera.
• Can be difficult to analyse.
• Provides rich, qualitative data.

CONTROLLED OBSERVATION

• Researchers control some variables.

Reliability – How consistent a measuring device is.

Inter-rater reliability - How closely the ratings of two observers compare.

Validity – Whether a study measures what it set out to.

Self-report

• Any method that involves asking the participant about feelings, attitudes, beliefs etc.

QUESTTIONAIRES

• Consist of a set of questions.
• Open questions - invite the respondent to provide their own answers.

-provide qualitative data.

-difficult to analyse.

-can produce more in-depth responses.

• Closed questions - provide a limited choice of answers.

-Provide quantitative data.

-Easy to analyse.

-Do not allow for in-depth insights.

• Likert scale - A scale used for participant to decide how strongly they agree or disagree with statements.