IB Psychology Paper 3

Mean

The mathematical average of a data set, calculated by adding all scores and dividing by the total count. It is highly sensitive to outliers, which distort the true central tendency.

Median

The middle value when a data set is ordered from lowest to highest. It ignores the specific numerical values of all other scores, making it less useful for further statistical testing.

Standard Deviation

• A measure of how spread out data points are from the average (mean).

• Large SD: High variability/spread —> data is less consistent (points are far from the mean).

• Small SD: Low variability/spread —> data is very consistent (points are clustered close to the mean).

Correlation Coefficient ($r$)

A numerical value between $-1.00$ and $+1.00$ indicating the strength and direction of a relationship between two variables, characterized by bidirectional ambiguity rather than causality.

Statistical Significance ($p$-value)

The probability that observed results occurred purely by random chance, typically considered acceptable when the probability is less than $0.05$. It does not indicate effect magnitude or practical importance.

Ceiling Effect

A measurement issue occurring when a test is too easy, causing scores to cluster at the maximum possible level and obscuring true participant variance.

Floor Effect

A measurement issue occurring when a test is too hard, causing scores to cluster at the minimum possible level and obscuring true participant variance.

Method Triangulation

The practice of combining multiple data-collection procedures (such as surveys and interviews) to cross-verify findings, eliminate bias, and enhance credibility.

Data Triangulation

Gathering evidence across different times, physical spaces, or participants to cross-verify findings and bolster research credibility.

Researcher Triangulation

Employing multiple independent observers or analysts to collect and evaluate evidence, minimizing individual subjective bias.

Personal Reflexivity

Critical self-examination regarding how an investigator's personal values, beliefs, and life experiences influence study design, execution, and interpretation.

Epistemological Reflexivity

Critical reflection on how foundational assumptions and the chosen research design constrain and shape study findings.

Reflexive Journal

A continuous record maintained by an investigator to actively identify and isolate subjective values from distorting data tracking and analysis.

Researcher Bias

A threat to credibility where an investigator's preconceived expectations unconsciously influence data collection or interpretation, often mitigated by double-blind procedures.

Participant Expectations

A demand characteristic where subjects alter their natural behavior or answers to align with what they believe the study seeks to find.

Social Desirability Bias

The tendency of respondents to distort answers to present a favorable image to others, minimized through complete anonymity.

Thick Description

Providing extensive, detailed accounts of the research context, participants, and interactions to enable comparison with other settings.

Transferability

The extent to which qualitative research findings can be applied to new contexts, made feasible only when rich descriptive detail of the original setting is provided.

Truncated Axis

A graphical distortion where the vertical axis begins at a number greater than $0$ (such as $60\%$), visually exaggerating minor differences between comparison groups.

Missing Error Bars

A visual flaw where charts show group averages without dispersion boundaries, hiding group overlap and obscuring whether differences are meaningful or variance-driven.

Unequal Interval Scaling

Inconsistent spacing between numerical increments on a graph axis, distorting the visible trajectory of a trend line and generating artificial spikes or plateaus.

Semi-structured vs. Focus Group Interviews

Interviews using flexible, open-ended prompts conducted with a single participant versus facilitated discussions among $6$ to $10$ participants that leverage interaction but risk conformity bias.

Participant vs. Non-participant Observation

A qualitative approach where the researcher directly embeds within the target group (risking loss of objectivity) versus watching events unfold from a detached distance without interaction.

Purposive Sampling

A non-probability recruitment technique that selects individuals who specifically match predefined criteria relevant to the research objective (e.g., patients diagnosed with PTSD).

Snowball Sampling

A recruitment method where current subjects refer their acquaintances to take part, specifically utilized for accessing marginalized, hidden, or hard-to-reach populations.

Credibility (Qualitative Research)

The extent to which research conclusions accurately reflect the true meanings and subjective experiences of the participants, acting as the qualitative equivalent of internal validity.

Member Checking

A trustworthiness procedure where transcripts, field notes, or analytical interpretations are presented back to participants to verify that their viewpoints were portrayed accurately.

Key Ethical Considerations (Paper 3)

Core methodological safeguards required in qualitative studies: informed consent, protection from physical and psychological harm, right to withdraw, confidentiality/anonymity, and comprehensive debriefing.