Experimental methods

What is an experiment

• A research method in which the researcher manipulates an independent variable (IV) to see if it causes a change in a dependent variable (DV), while controlling other variables.

• The key idea: to establish cause and effect (causal inference). Seneca Learning+1

• Must operationalise variables (i.e. define them in measurable, concrete terms) so they can be manipulated or measured

Why experiments are used in psychology

• To test hypotheses under controlled conditions.

• They allow controlling extraneous variables, reducing confounding influences.

• Allow replication (if procedures are clearly defined).

• Strong internal validity (when done well).

• But must trade off with ecological validity sometimes.

lab experiments

• conducted in tightly controlled environments

• the experimenter deliberately manipulates the IV across conditions

• experiment measures the dependent variable - produces quantitative data

• Experimenter controls extraneous variables 

• strengths:

• procedure and instructions are standardized

• high control over variables and standardized - strong internal validity, easier replication, clear operationalization

• Extraneous and confounding variables are minimized.

• The researcher can confidently claim cause and effect between IV and DV.

• Easier replication

  • Standardized procedures make it easy for other researchers to repeat the study → test reliability.

• Statistical analysis is straightforward

  • Quantitative, numerical data means easier statistical testing and clear comparisons between conditions.

weakness:

Low ecological validity

• The artificial setting may not reflect real-life behavior.

• Participants might not behave naturally in lab environments.

• Demand characteristics

  • Participants guess the study’s aim and alter their behaviour (trying to help or sabotage results).

  • Low mundane realism

    • Tasks are often artificial (e.g., memorizing nonsense syllables, pressing buttons).

    • Findings may not be applicable to complex real-world behavior.

  • Experimenter effects

    • Subtle cues from the researcher (tone, body language) can unconsciously influence participants’ responses.

    • Ethical issues

      • The need for control may lead to deception or stress (e.g. Milgram’s obedience study).

field experiments

• are conducted in a more natural environment

• The experimenter deliberately changes the independent variable

• The experimenter measures the dependent variable - quantitative data

• the experimenter controls some of the extraneous variables

• strengths:

•  Higher ecological validity - Behaviour occurs in natural settings → results are more generalizable to real life - the results are more accurate

• Reduced demand characteristics

  • Participants are often unaware they are in a study, so behaviour is more genuine (high mundane realism).

• Useful for studying real-world behaviour

  • Can explore social influence, obedience, helping behaviour, etc. in natural contexts.

• Good balance between control and realism

  • Some control is possible while maintaining natural settings.

• Practical applications

  • Findings can be used to improve real-world outcomes (e.g., reducing littering, promoting prosocial behaviour).

weakness:

• Less control over extraneous variables

  • Environmental factors (weather, noise, bystanders) can affect the DV, reducing internal validity.

• Replication is difficult

  • Because real-world settings vary, it’s hard to repeat exactly → reliability decreases.

• Ethical concerns

  • Participants often can’t give informed consent or may be deceived (since they don’t know they’re being studied).

• Practical difficulties

  • Gaining access to settings, coordinating manipulations, and collecting data covertly can be challenging.

• Observer bias

  • If the researcher observes directly, subjective interpretation may influence results.

• Confounding variables

  • Harder to isolate the IV → causality less certain.

natural experiment

• the experimenters have no control of the IV

• The experimenter measures the DV

• The experimenter has no control over extraneous variables

strengths:

• ✅ Strengths

  1. Allows study of ethically or practically impossible variables

     • You can study phenomena like the effects of stress, trauma, or illness without creating them artificially.

• High ecological validity

  • Since the event occurs naturally, behavior tends to reflect real life.

• Opportunities for unique insights

  • Explores naturally occurring changes that can’t be replicated (e.g., before-and-after disaster studies).

• Less experimenter interference

  • Reduces risk of demand characteristics or experimenter bias (IV not manipulated by researcher).

• Can provide evidence for causal relationships

  • If confounding variables are minimized, results can still suggest causation (though less strongly than in labs).

weakness:

Lack of control over variables

• Confounding variables can’t be easily ruled out → lower internal validity.

• Random allocation not possible

  • Participants fall into groups naturally → increases risk of participant variables affecting results.

• Replication is difficult

  • Natural events rarely repeat in exactly the same way.

• Ethical issues

  • Studying sensitive situations (e.g., trauma, illness) may raise issues of privacy and psychological harm

• Low control over data collection

  • Natural events may occur suddenly; researchers must use opportunistic sampling or retrospective methods, which can reduce reliability.

Qusai-experiments

• There is an independent variable (IV) and a dependent variable (DV).

• BUT the researcher does not randomly assign participants to the conditions.

• The IV is a pre-existing characteristic — something that already exists and cannot be manipulated by the researcher.

• strengths:

• Ethically possible

  • Lets psychologists study variables that can’t be manipulated ethically (e.g., gender, mental illness, age).

• Often done under controlled conditions

  • Many quasi-experiments take place in labs → high control → fairly high internal validity.

• Allows comparison between different groups

  • Reveals how natural characteristics affect behaviour.

• Useful for understanding real-world differences

  • Helps understand the effects of disorders, conditions, or social factors.

weakness:

1. Causality is uncertain

   • Because the researcher doesn’t manipulate the IV, we can’t say one thing caused another — only that they’re linked.

2. Limited control over participant variables

   • The groups (e.g., men vs women) may differ in other uncontrolled traits (e.g., upbringing, education).

3. Harder to generalize

   • Often use small, specific samples (e.g., people with a rare condition).

4. Sometimes, low ecological validity

   • If conducted in a lab setting, the tasks might still be artificial.