IB Psychology - Research methods

Quantitative Research

Research where data comes in numbers, aiming to derive universally applicable laws (nomothetic approach).

Qualitative Research

Research focusing on in-depth study of phenomena through texts, aiming for understanding meanings (idiographic approach).

Construct

Any theoretically defined, unobservable variable such as aggression, memory, or anxiety.

Experiment

A quantitative method involving one independent variable (IV) and dependent variable (DV) with controlled extraneous variables to infer cause and effect.

Correlational Study

A quantitative study where no variables are manipulated, and the relationship between measured variables is quantified.

Population Validity

The extent to which experimental results can be generalized from the sample to the target population.

Ecological Validity

The extent to which the results of an experiment may be applied to natural real-life settings.

Construct Validity

The extent to which a study measures the hypothesized theoretical construct rather than something else.

Internal Validity

The experimental measure indicating whether manipulation of the IV caused the observable change in the DV.

Random Sampling

A sampling technique where every member of the target population has an equal chance of becoming part of the sample.

Experimenter Bias

When the researcher unintentionally exerts influence on the results of the study.

Double-Blind Design

A procedure where information that could introduce bias is withheld from both participants and experimenters.

Quasi-Experiment

A study with controlled group allocation that lacks complete randomization, limiting absolute cause-and-effect inferences.

Natural Experiment

A quasi-experiment where experimental manipulation occurs naturally without researcher intervention.

Laboratory Experiment

An experiment conducted in a specially designated venue with high internal validity but potentially lower ecological validity.

Field Experiment

An experiment conducted in a real-life setting, offering higher ecological validity but less control over confounding variables.

The environment is not controlled

Variable

a construct - any characteristic, factor, or quantity in a study that can be changed, controlled, measured, or counted

Independent variable

The experimental factor that is manipulated; the variable whose effect is being studied.

Dependent variable

The outcome factor that is measured; the variable that may change in response to manipulations of the independent variable.

Causality (Cause-and-Effect)

a relationship where a change in one variable—the independent variable (IV)—directly produces a change in another variable—the dependent variable (DV)

Causality vs correlation

All causal relations are correlations. Not all correlations are causal relations.

The most critical rule in research is that correlation does not imply causation. Even if two variables are strongly linked, you cannot conclude that one influences the other based on a correlational study alone

Treatment group vs control group

Treatment group: The group of participants that receives the deliberate manipulation of the independent variable (the "treatment" or intervention) being tested by the researcher

Control group: The baseline group of participants that does not receive the experimental treatment. Instead, they may receive no treatment, a standard baseline condition, or a placebo (a fake treatment)

Remember that participants should be randomly assigned to either the treatment or control group to keep extraneous variables balanced and ensure high internal validity

Extraneous variable

In an experiment, a variable other than the IV that might cause unwanted changes in the DV.

Confounding variable

a factor other than the independent variable that might produce an effect in an experiment

extraneous variable vs confounding variable

all confounding variables are extraneous, but not all extraneous variables are confounding

An extraneous variable that varies systematically with the independent variable, providing an alternative explanation for the results (It affects the dependent variable and is systematically related to (or covaries with) your independent variable)

Controls

constraints that the experimenter places on the experiment to ensure that each subject has the exact same conditions, except the manipulation of EV

True experiment (laboratory experiment)

1. Participants are randomly allocated to the different experimental conditions or groups

2. EVs are controlled for, IV is isolated as the only factor affecting the DV

3. In a controlled experiment

Correlation

A measure of the extent to which two factors vary together, and thus of how well either factor predicts the other.

Bidirectional ambiguity

a limitation of correlational research where a statistical relationship is established between two variables, but it is impossible to determine the direction of causality.

How and why are true experiments used?

"How" (+ purpose)

• Random allocation: Distributes participant variables evenly across groups to eliminate selection bias

• Controlled environment: Ensures that extraneous variables do not interfere with the data.

• EVs are controlled for so that IV is the only factor that affects DV

+ use of control group: to verify if changes in the DV are truly caused by the IV rather than external factors

“Why”

• High internal validity: By strictly controlling confounding variables, researchers can confidently state that changes in the DV were solely produced by the IV.

• Replicability: The standardized, structured nature of the method allows future researchers to repeat the exact same procedure

Evaluation

• Strength

  • Minimizes confounding variables through tight environmental controls

• Limitation

  • Low Ecological Validity: Laboratory settings create highly artificial environments, meaning results may not generalize to real life

  • Participant expectancy effect: Participants may guess the aim of the study due to the controlled setting and subconsciously alter their natural behavior to match expectations.

+ It is always better to give an example case study

Loftus and Palmer (1974)

• Aim: To investigate if leading questions could systematically alter speed estimates in eyewitness testimony.

• Design: Laboratory experiment using an independent measures design. 45 students watched clips of traffic accidents.

• Random Allocation: Participants were randomly split into 5 distinct conditions (9 per group).

• IV Manipulation: The specific verb used in the critical question: "About how fast were the cars going when they [smashed / collided / bumped / hit / contacted] each other?"

• DV Measurement: The numerical speed estimate provided by the participant in miles per hour (mph).

• Controls: All participants viewed the exact same video clips under the exact same structural layout.

• Findings: The verb "smashed" yielded the highest mean estimate (40.5 mph), while "contacted" yielded the lowest (31.8 mph).

• Explicit Link to Question: Because this was a true experiment with strict isolation of the verb (IV), the researchers empirically proved that post-event information caused a direct shift in cognitive memory reconstruction (DV).

Independent samples design (independent groups design)

An experimental design where different participants are allocated to each condition of the Independent Variable (IV)

Strength and Limitation of Independent samples

Strength

• Random allocation control EVs:

  • Participant expectancy effect (participants suspect results and alter their behavior)

  • Order effects (the sequence of experimental conditions influences participant performance rather than the IV)

Limitation

• Participant variability: natural differences in participants could influence the results

Repeated measures

All participants experience all conditions of the experiment

Strength and limitation of repeated measures

Strength

• Limit participant variability

• Do not require as many participants to run an experiment

Limitation

• Order effects: can be controlled through counterbalancing (randomizing the order in which conditions are presented)

• Higher chance for participant expectancy effect

Matched pairs

Participants are matched on relevent criteria and then allocated to different conditions

Strength and limitation of matched pairs

Strength

• limits participant variability: by matching participants on key variables that could affect the outcome, ensure the groups are highly comparable

Limitation

• may affect the value of randomization & potential risk of selection bias

• difficult & time-consuming: finding each pair who match perfectly on specific traits requires massive amounts of pre-testing and filtering

• Impossible to match perfectly: hidden participant variability can still affect the results

Researcher bias and how to control it

happens when a researcher's own personal beliefs, expectations, values, or preferences unintentionally influence how a study is designed, conducted, or interpreted

can be controlled through;

• Single-blind design: When participant does not know which group they belong to

  • low implementation complexity: simple to organize and manage

  • still vulnerable to researcher expectations

• Double-blind design: When neither the participants nor the researchers know which group they belong to

  • high implementation complexity; requires a third party to handle data coding and tracking

  • high scientific rigor; useful for objective research

What is the overall benefits of experimental designs & controls?

Increases the internal validity (the extent to which the ecidence found actually supports a cause-effect claim; the extent to which the change in the DV is due to the manipulation of the IV)