Types of Experiment (College Board AP® Psychology): Revision Note

Laboratory experiments

• Psychological research can be broadly divided into experimental and non-experimental methodology:

  • Experimental methodology involves the manipulation of an IV to observe its effect on a DV

  • Non-experimental methodology does not involve manipulation of an IV — the researcher observes, measures or records variables as they naturally occur

• A lab experiment is a type of experimental method in which the researcher exerts high levels of control over the procedure

• The researcher manipulates the IV and measures its effect on the DV under controlled conditions

• Environmental factors that could act as confounding variables (e.g. noise, temperature, lighting) are controlled to ensure that any change in the DV can be attributed to the IV

• A standardized procedure is used to ensure replicability and reliability:

  • The experiment is run in the same environment across all conditions

  • All participants receive the same instructions

  • Only the IV changes between conditions

• Lab experiments produce quantitative data — numerical, objective measurements of the DV (e.g. reaction time in milliseconds, number of words recalled out of 20)

Evaluation of lab experiments

Strengths

• It is easier to establish a cause-effect relationship between the IV and the DV than for other methods used in psychology

  • This is due to the use of controls and the objective nature of the research

  • Therefore lab experiments are high in internal validity

• The use of a standardized procedure means the study can be replicated, which allows reliability to be checked

Limitations

• The use of artificial tasks means that lab experiments lack validity

  • If participants are performing tasks in an unfamiliar, 'sterile' setting, this does not reflect how they might behave in real conditions

  • This makes the findings difficult to generalize beyond the lab setting

• Demand characteristics may limit the generalizability of the findings

  • As participants know they are being studied, they may alter their behavior based on what they think is expected, reducing the validity of the findings

Field experiments

• A field experiment is a type of experimental method conducted in a natural, real-world setting

• The researcher still manipulates the IV and measures the DV

  • This distinguishes field experiments from naturalistic observation

• For example:

  • A confederate pretends to collapse on a subway train

  • IV: whether the confederate appears drunk or disabled

  • DV: the number of people who go to the confederate's aid

• The researcher has less control over the environment than in a lab, meaning extraneous variables are more likely to be present, e.g.

  • weather conditions affecting participants' mood

  • noise, crowds, or other environmental distractions

• Field experiments produce quantitative data as the primary measure, e.g.

  • the number of people who help a confederate

  • scores on a behavioral questionnaire

Evaluation of field experiments

Strengths

• As the research is conducted in real-world settings, participant behaviour is more likely to reflect their natural behavior outside of the study

  • This makes field experiments high in validity

• Participants are less likely to experience demand characteristics, particularly if they are unaware they are taking part in a study, which increases the validity of the findings

Limitations

• Extraneous variables are much harder to control in a natural setting

  • This reduces internal validity and makes it more difficult to establish a clear cause-and-effect relationship

• Field experiments are more difficult to replicate than lab experiments due to the uncontrolled nature of the setting

  • This reduces reliability

Natural experiments

• A natural experiment is a non-experimental method — the researcher does not manipulate the IV

  • The IV occurs naturally, meaning the researcher cannot randomly assign participants to conditions

• The researcher identifies a naturally occurring event or condition and measures its effect on a DV of their choosing, e.g.

  • the effect of living through a natural disaster on stress levels

  • the effect of experiencing a specific mental illness on cognitive performance

• Because the IV cannot be manipulated, naturally occurring phenomena are often studied — these would be unethical to impose on participants deliberately

• Natural experiments typically produce quantitative data where the DV is objectively measured

  • E.g. standardized stress scores, cognitive test scores

• However, unlike quasi-experiments, the researcher has limited control over the procedure as a whole — not just the IV

Evaluation of natural experiments

Strengths

• They allow the researcher to investigate topics which would be unethical to study using a lab experiment e.g. experiencing a mental illness or a natural disaster

  • This means that natural experiments are high in validity

• Participants report on real, first-hand experiences and the researcher does not attempt to control the procedure

  • This means that natural experiments are high in ecological validity

Limitations

• Causal relationships between the IV and DV are difficult to determine due to the number of uncontrolled variables

  • This is a key limitation of a natural experiment as it imposes no controls on the procedure

• The researcher cannot be certain what effect the naturally occurring phenomena have had on participants

  • This reduces the reliability of the findings

• Natural experiments may be affected by sample bias — only people who have experienced the specific phenomenon can participate, limiting generalization

Quasi experiments

• A quasi-experiment is a non-experimental method — the researcher does not manipulate the IV

  • The IV is a pre-existing characteristic of the participants rather than something imposed by the researcher e.g.

    • age (young vs. older participants)

    • gender (female vs. male participants)

    • experience (trained vs. untrained participants)

• Because the IV is a participant characteristic, participants cannot be randomly assigned to conditions — they are already in a condition by definition

• Quasi-experiments produce quantitative data and can follow a similar procedure to a true lab experiment

  • The only variable the researcher cannot control is the IV

• For example:

  • IV: whether participants are young (18–25) or older (65+)

  • DV: number of digits correctly recalled on a digit-span task

Evaluation of quasi experiments

Strengths

• Because the IV is a real characteristic rather than an artificially imposed condition, the results may be higher in external validity and more generalizable to real-world populations

• Quasi-experiments follow a structured experimental design and can be replicated with participants who match the original sample in terms of demographics

  • E.g. the effect of age on recall could use the same procedure over and over again

Limitations

• The inability to randomly assign participants to conditions means participant variables may act as confounding variables, making it difficult to determine causality

  • E.g. a study investigating the effect of age on recall might include a group of participants who naturally have a much better memory than is representative of their age group, skewing the results

    • This reduces the internal validity of the findings

• Quasi-experiments lack internal validity because factors beyond the IV could explain the results

  • E.g. teachers trained in empathy may work in schools that already value emotional intelligence, meaning any advantage on an emotion-recognition task could reflect their environment rather than the training itself