PaaS- non experimental methods

Non-experimental research

Quantitative research that measures variables without manipulating them or randomly assigning participants to conditions.

Correlational design

Measures two or more variables to examine their relationship without inferring causation.

Why non-experimental methods are used

They are used when manipulation is unethical, impractical, or when researchers are interested in relationships rather than causes.

Key limitation of correlational research

It cannot determine cause and effect because alternative explanations cannot be ruled out.

Quasi-experiment

A study that resembles an experiment but lacks random allocation to conditions.

Why random allocation matters

Random allocation reduces individual differences between groups and increases internal validity.

Pre-post design

Measuring the same participants before and after an intervention to detect change.

Selection bias

Differences between groups that exist before the study begins, which can influence results.

Attrition bias

When dropouts differ systematically between groups, distorting results.

History and maturation effects

External events or natural changes over time that may explain observed differences.

Natural experiment

A quasi-experiment where group differences occur naturally due to external events or policies.

Strength of natural experiments

They allow real-world phenomena to be studied with high ecological validity.

Limitation of natural experiments

Confounding variables cannot be controlled, so causation cannot be inferred.

Survey research

Collecting self-report data to describe or relate psychological variables.

Cross-sectional survey

Measures variables at a single point in time.

Longitudinal survey

Measures the same individuals over time to examine change.

Social desirability bias

Tendency to give answers that appear socially acceptable rather than truthful.

Recall bias

Inaccurate memory of past behaviours or experiences.

Systematic observation

Structured recording of behaviour in natural or controlled settings.

Secondary data analysis

Using existing datasets collected by other researchers.

Systematic review

A structured summary of all available evidence on a research question.

Meta-analysis

Statistically combining results from multiple studies to estimate an overall effect.

Publication bias

The tendency for studies with significant results to be published more than null findings.

Design trade-offs

Non-experimental methods trade causal certainty for realism and feasibility.

Key evaluation rule

Non-experimental studies must not be used to make causal claims.