Learning Outcomes

• Define key research concepts: population, sample, sampling frame, generalisation, bias.
• Differentiate major sampling methods in health and dental research.
• Explain the influence of sampling on validity and generalisability.
• Identify common bias sources in oral health studies.
• Assess quality of oral therapy research.
• Interpret findings with an understanding of sampling and bias effects.

Sampling Concepts

• Sample: Group representing a larger population in research.
• Population vs Sample:
  • Population: Entire group of interest.
  • Sample: Subset studied.

Importance of Sampling

• Impractical to study entire population.
• Well-chosen samples enable inference to the whole population.
• Sample representativeness affects trustworthiness of findings.

Population Types

• Target Population: Ideal group for generalisation.
• Accessible Population: Available group for recruitment; often differs from target due to limitations (time, cost, ethics).

Sampling Methods

• Probability Sampling: Random selection; unbiased samples; allows generalisation.
• Non-Probability Sampling: Non-random selection; practical but may introduce bias.

Types of Probability Sampling

• Simple Random Sampling: Equal chance for all individuals.
• Systematic Sampling: Selected at regular intervals.
• Stratified Sampling: Divided into subgroups, random samples from each.
• Cluster Sampling: Entire groups are selected first.

Types of Non-Probability Sampling

• Convenience Sampling: Easiest participants recruited.
• Quota Sampling: Pre-set categories filled non-randomly.
• Snowball Sampling: Participants recruit others from their network.

Sample Size Considerations

• Too small: Unreliable results; risk of missing differences.
• Too large: Unnecessary costs and burden.
• Sample size influenced by:
  • Research design.
  • Expected effect size.
  • Statistical power.
  • Population variability.

Generalizability (External Validity)

• Extent findings can apply to other settings/groups.
• Generalizable when sample accurately represents the population.
• Affected by biased samples and narrow criteria.

Bias in Research

• Bias: Systematic error distorting findings.
• Types of Bias:
  • Sampling Error: Method inadequacy alters representativeness.
  • Selection Bias: Systematic manipulation during enrollment.
  • Allocation Bias: Differences in participant group characteristics.
  • Observer Bias: Researcher expectations affect measurements.
  • Recall Bias: Participants misremember past events.
  • Reporting Bias: Selective reporting of positive behaviours.

Mitigating Bias

• Techniques to reduce bias include randomisation, blinding, standardised data collection, and ensuring representative samples.