TYPES OF RESEARCH METHODS:
AO1 1.) content analysis
Qualitative to Quantitative
* Quantify Communication Patterns
* Systematic Sampling
* Identify Coding Units
* Operational Definitions
* Tally Occurrences
AO3 content analysis
✖ Observer interpretation bias
✖ Coding subjective
✔ Inter-rater reliability checks
✔ Improves consistency
✔ Increases data confidence
AO1 Thematic analysis
Remains Qualitative
* Transcribe Data
* Identify Themes/Trends
* Shorthand Coding
* Supporting Evidence/Quotes
AO3 Thematic analysis
✔ High ecological validity
✔ Real-life qualitative data
✖ Unscientific process
AO1 2.) Case studies
Detailed/In-depth
* Unusual/Rare Cases
* Multi-method Approach
* Range of Sources
* Identify Themes
* Longitudinal Study
AO3 case studies
✔ Rich detailed data
✔ In-depth insight
✖ Ethical issues
✖ Researcher involvement
✖ Low generalisability
AO1 - Reliability
- Results Consistency
- Test-retest Method
- Inter-observer Reliability
- +0.8 Correlation
- Standardized Procedures
- Operationalized Categories
AO3 - Reliability
- Lab Experiments (High)
- Unstructured Interviews (Low)
- Recall/Attitude Bias
AO1 - Validity
- Results Legitimacy
- Internal/External Validity
- Ecological/Temporal Validity
- Face Validity (“Eyeballing”)
- Concurrent Validity
- Single/Double Blind
AO3 - Validity
- Demand Characteristics
- investigator Effects
- Covert Observation (High)
AO1 - Features of Science
- Empirical Methods
- Objectivity
- Replicability
- Falsifiability
- Theory Construction
- Paradigm/Paradigm Shifts
AO3 - Features of Science
- Pseudoscience (Non-falsifiable)
- Controlled Environments
- Hypothesis Testing
AO1 - Improving Reliability
- Rewrite/deselect ambiguous questions
- Replace open with closed questions
- Same interviewer/proper training
- Use structured interviews
- Standardized procedures/instructions
- Operationalize behavioral categories
AO3 - Improving Reliability
- Lab experiments high control
- Unstructured interviews low reliability
- Measurable/self-evident checklist behaviors
AO1 - Improving Validity
- Use of control groups
- Single/double-blind procedures
- Lie scales for social desirability
- Anonymity for respondents
- Covert observations
- Clearly operationalized categories
AO3 - Improving Validity
- Reduced demand characteristics
- Natural behavior in covert
- Minimized investigator effects
AO1 - Reporting Investigations
- Abstract (summary)
- Introduction (background/aims)
- Method (procedure/participants)
- Results (findings/stats)
- Discussion (interpretation/limitations)
- Referencing (sources)
AO3 - Reporting Investigations
- Scientific credibility
- Allows for replication
- peer review/objectivity
AO1 - Levels of Measurement
- Nominal (categories/tallies)
- Ordinal (ordered/ranked)
- Interval (fixed/universal scale)
- Ratio (absolute zero)
- N-O-I mnemonic
- Parametric vs non-parametric
AO3 - Levels of Measurement
- Ordinal lacks universal scale
- Questionnaire data usually ordinal
- Interval most precise/sophisticated
AO1 - Probability & Significance
- Calculated vs Critical values
- Significance level (P≤0.05)
- 5% probability of chance
- Type I error (optimistic/false positive)
- Type II error (pessimistic/false negative)
- One-tailed vs Two-tailed tests
AO3 - Probability & Significance
- 5% avoids both errors
- 1% level for drug trials
- Statistical validity/objectivity
AO1 - Statistical Tests
- Spearman’s rho (Correlation/Ordinal)
- Pearson’s r (Correlation/Interval)
Wilcoxon (Difference/Related/Ordinal) - Mann-Whitney (Difference/Unrelated/Ordinal)
- Related t-test (Difference/Related/Interval)
- Chi-Squared (Difference or Association/Nominal)
AO3 - Statistical Tests
- Parametric tests more accurate
- Rule of R (calculated ≥ critical)
- “Carrots Should Come…” mnemonic
