Psychology key science skills

Aim

To examine the effect of caffeine on short-term memory recall. A precise statement of what the study intends to investigate.

Hypothesis

A testable prediction including: population, IV, DV, and direction (e.g., increase/decrease).
‎
Example: "Year 12 students who consume caffeine will recall more words than those who do not."

Independent Variable (IV)

The variable manipulated by the researcher.

Dependent Variable (DV)

The outcome measured and affected by the IV.

Controlled Variables

Factors kept constant to ensure validity.

Confounding Variables

Uncontrolled variables that affect the DV and may invalidate results.

Within-Subjects Design (Repeated Measures)

Same participants do all conditions. Controls for participant differences but may cause order effects.
‎

Counterbalancing helps minimise these.

Between-Subjects Design (Independent Groups)

Participants are assigned to one condition only. No order effects but can suffer from participant differences. Random allocation helps reduce this bias.

Mixed Design

Combines within- and between-subjects methods. Useful for tracking individual changes over time and group differences.

Convenience Sampling

Selects participants who are easily available.
‎
✅ Quick and easy.
❌ High bias; low external validity.

Random Sampling

Everyone in the population has an equal chance of being selected.

‎

✅ Reduces bias; improves generalisability.

Stratified Sampling

Divides population into strata (e.g. age, gender) and samples proportionally.

‎

✅ Highly representative; allows for group comparisons.

Random Allocation

Each participant has an equal chance of being in any condition.

‎

✅ Controls participant variables and supports internal validity.

Random Errors

Unpredictable variations (e.g. mood, distractions).

Systematic Errors

Consistent bias due to issues like faulty equipment or poor sampling.

Personal Errors

Human mistakes like incorrect data recording.

Extraneous Variables

Other variables that might influence the DV. Must be controlled to maintain internal validity.

Confounding Variables (again)

Uncontrolled variables that impact the DV and can invalidate the study.

Standardised Instructions

Ensure every participant gets the same procedure to reduce bias.

Counterbalancing

Changing the order of conditions to reduce order effects like practice or fatigue.

Single Blind Procedure

Participant doesn't know which condition they're in.

Double Blind Procedure

Both participant and researcher are unaware of condition assignments.

Triple Blind Procedure

Participant, researcher, and data analysts are unaware of group allocations.

Case Study

In-depth study of one person or group.

‎

✅ Rich detail.

❌ Low generalisability; no causation inferred.

Observational Study

Watching behaviour naturally or in a set-up.

‎

✅ Realistic.

❌ May lack consent; observer bias possible.

Self-Reports

Surveys, questionnaires, or interviews.

‎

✅ Efficient; good for subjective data.

❌ Prone to response bias.

Correlational Study

Measures strength/direction of a relationship.

‎

✅ Good for ethically tricky studies.

❌ No causation inferred.

Simulation Study

Realistic re-creation of scenarios.

‎

✅ Safe and controlled.

❌ May lack ecological validity.

Primary Data

Data collected directly by the researcher.

Secondary Data

Pre-existing data (e.g. ABS, other studies).

Quantitative Data

Numerical data – easy to analyse statistically.

Qualitative Data

Descriptive data – thoughts, feelings, behaviours.

Objective Data

Observable and measurable (e.g. reaction time). Not influenced by personal feelings.

Subjective Data

Based on personal opinions or self-reports.

Reliability

Consistency of results over time, observers, or tools.

Repeatability

Same method, same conditions, same results.

Reproducibility

Different researchers or settings get similar results.

Validity

The degree to which something measures what it's supposed to.

Internal Validity

Confidence that the IV caused changes in the DV.

External Validity

Generalisation of results to real-world settings.

Research Merit and Integrity

Study must be worthwhile, well-designed, and honestly reported.

Beneficence

Benefits must outweigh risks.

Justice

Fair distribution of risks and benefits among participants.

Respect

Participant rights and dignity must be upheld.

Confidentiality

Participants’ data must be protected and kept private.

Voluntary Participation

Participants must willingly choose to take part.

Informed Consent

Participants must be fully informed before agreeing to take part.

Withdrawal Rights

Participants can leave the study at any time without penalty.

Protection from Harm

Participants must not be exposed to any risk or harm.

Deception (with Debriefing)

If deception is used, a full explanation must be provided afterwards.