Debates

Extraneous Variables

Any variable other than the IV that affects the DV.

❌ Reduces internal validity by introducing alternative explanations.

✅ Standardization, randomization, counterbalancing.

Confounding Variables

An extraneous variable that systematically affects one condition more than another.

❌ Directly threatens validity, making results unreliable.

✅ Matched participants, controlling for prior exposure.

Situational Variables

Aspects of the environment that may influence behavior (e.g., noise, lighting, temperature).

❌ Can differ across conditions, affecting results.

✅ Keep environmental factors constant across conditions.

Participant Variables

Individual differences between participants (e.g., intelligence, mood, motivation).

❌ Affects DV independently of IV, reducing reliability.

✅ Matched Pairs Design, Repeated Measures, Random Allocation.

Standardization

Keeping procedure, materials, and instructions the same for all participants.

✅ Eliminates procedural inconsistencies.

Randomization

Using random allocation to assign participants to different conditions.

✅ Reduces participant variables.

Counterbalancing

Alternating the order of conditions in Repeated Measures Design.

✅ Controls for order effects (practice/fatigue).

Single-Blind Design

Participants do not know which condition they are in.

✅ Reduces demand characteristics.

Double-Blind Design

Both participants and experimenters do not know the condition assignments.

✅ Prevents researcher bias.

Pilot Study

A small-scale trial before the main experiment.

✅ Identifies uncontrolled variables early.

Type

Definition

Key Considerations

Internal Validity

The extent to which the study controls confounding variables, ensuring that changes in the DV are caused by the IV and not extraneous factors.

✅ High if: Strong controls, no demand characteristics. ❌ Low if: Confounding variables present, participant bias.

Ecological Validity

The extent to which findings generalize to real-world settings.

✅ High if: Natural setting, realistic tasks. ❌ Low if: Artificial lab environment, unnatural behavior.

Mundane Realism

The degree to which the tasks in a study resemble real-life situations.

✅ High if: Task is something participants normally do. ❌ Low if: Unfamiliar, artificial tasks (e.g., memorizing nonsense words).

Face Validity

The extent to which a test appears to measure what it claims to measure (superficial assessment).

✅ Quick initial check for validity. ❌ Does not guarantee scientific accuracy.

Concurrent Validity

The extent to which a new test correlates with an established, validated measure.

✅ High if: Strong correlation with existing test results. ❌ Low if: Poor correlation with validated measures.

Generalizability

The extent to which study findings apply to different populations, settings, and time periods.

✅ High if: Large, diverse sample. ❌ Low if: Small, biased sample (e.g., only students).

Threat

Description

Solution

Demand Characteristics

Participants alter behavior based on perceived study aims.

❌ Reduces validity. ✅ Use deception, single-blind design.

Researcher Bias

Researcher’s expectations influence participant behavior or result interpretation.

❌ Lowers objectivity. ✅ Use double-blind design, independent coders.

Extraneous Variables

Uncontrolled factors affect the DV outside of the IV’s influence.

❌ Introduces alternative explanations. ✅ Use standardized procedures, control conditions.

Social Desirability Bias

Participants respond in a way that appears favorable rather than truthfully.

❌ Affects self-report methods. ✅ Assure anonymity, use indirect questioning.

Type

Definition

Key Considerations

Internal Reliability

Consistency within a study – Are all participants experiencing the same standardized conditions?

✅ High if: Fixed procedures, identical materials. ❌ Low if: Instructions vary, inconsistent tasks.

External Reliability

Consistency over time – Can the study be replicated and yield the same results?

✅ High if: Study is repeatable with the same outcome. ❌ Low if: Results fluctuate between trials.

Inter-Rater Reliability

The degree to which two or more observers produce the same qualitative data from the same event.

✅ High if: Clear coding system, trained observers. ❌ Low if: Observer bias, subjective interpretation.

Method

Definition

Application

Split-Half Method

Compare results from two halves of a test/questionnaire to check for internal consistency.

✅ High if: Two halves produce similar results. ❌ Low if: Different halves give inconsistent responses.

Test-Retest Method

Administering the same test twice to the same participants at different times and comparing results.

✅ High if: Scores remain stable over time. ❌ Low if: Results fluctuate significantly.

Aspect

Validity

Reliability

Definition

Measures accuracy – Does the study test what it claims to test?

Measures consistency – Are results stable over time?

Improved by:

✅ Controlling confounding variables ✅ Using realistic settings (high ecological validity)

✅ Standardizing procedures ✅ Using inter-rater reliability measures

Fails if:

❌ Study lacks control → Extraneous variables affect results.

❌ Study lacks replication → Results fluctuate.

Privacy

Avoids invasion of personal space or information disclosure.

✅ No unauthorized access to personal data.

Debriefing

Full explanation of study aims & consequences after participation.

✅ Ensures participants leave in same psychological state as they arrived.

Protection from Harm

No participant should experience greater physical or psychological risk than in daily life.

❌ Low protection → Ethical violation.

Informed Consent

Participants must receive sufficient details before agreeing to take part.

✅ Must be voluntary, with parental consent for under-16s.

Right to Withdraw

Participants must be able to leave at any time and request data removal.

✅ Clearly communicated before and during study.

Deception

Deliberate misinformation about the study is unethical unless justified.

✅ If necessary, must minimize distress and provide full debriefing.

Confidentiality

Participant identity & data must be kept secure and anonymous.

❌ Breach → Ethical misconduct.

Replacement

Consider alternatives (videos, simulations) instead of live animals.

✅ Reduces unnecessary animal use.

Species & Strain

Select a scientifically appropriate species.

✅ Consider previous experience & welfare.

Minimum Numbers

Use smallest number of animals to achieve research goals.

✅ Pilot studies/statistical calculations to justify sample size.

Procedures

Minimize stress, discomfort, or injury – requires Project Licence.

❌ Harm must pass cost-benefit analysis.

Pain & Distress

Avoid suffering; monitor and intervene if needed.

✅ Immediate action required for adverse effects.

Housing

Avoid isolation/crowding, ensure species-appropriate environments.

✅ Must reflect natural habitat and social behavior.

Reward & Deprivation

Food/water deprivation only if essential to research aims.

❌ No excessive deprivation → Must be ethically justified.

✅ Improves real-world applications (e.g., mental health, education, workplace efficiency).

❌ Some studies may be unethical in order to gain more valid results.

✅ Research findings can positively influence public policies.

❌ Low ecological validity in lab studies may limit real-world application.

✅ Helps identify whether behaviors are caused by personality (individual) or external environment (situational).

❌ Difficult to separate individual & situational influences, as they often interact.

✅ Understanding both perspectives allows for more effective interventions (e.g., therapy, policy changes).

❌ High ecological validity is needed to be useful in real-world contexts, but lab studies may limit this.

✅ Helps in understanding behavior origins (e.g., intelligence, aggression).

❌ Difficult to isolate nature/nurture effects, as they interact.

✅ Interactionist approaches (e.g., epigenetics) explain behavior more comprehensively.

❌ Nature-based explanations can be socially sensitive (e.g., genetic determinism in intelligence).

✅ Findings contribute to therapeutic interventions & education policies.

❌ Some studies may involve ethical concerns (e.g., twin studies, deprivation experiments).

Informed Consent

Must be obtained from parents/guardians.

Right to Withdraw

Children must be made aware of their right to leave at any time.

Protection from Harm

Studies must minimize psychological & physical distress.

Debriefing

Must ensure clear explanation in child-friendly terms.

Risk Assessment

A pre-study risk analysis must be conducted.