Evidence-Based Treatments and Research Methods

Evidence-Based Treatments and Research Methods

• Evidence-based treatments are therapies supported by scientific research.
• Critical consumers of research should evaluate the source of information, the evidence supporting claims, and potential biases.
• Empiricism relies on sensory evidence to draw conclusions.
• Quantitative methods use numerical data for analysis; qualitative methods use descriptive details.
• The theory-data cycle involves forming questions, making predictions, collecting data, and reflecting on findings.
• A theory is a set of statements explaining variable relationships; a hypothesis predicts research outcomes based on theory.
• Data are observations; a good theory is supported by data, falsifiable, and parsimonious.
• Applied research aims to solve practical problems; basic research expands knowledge.
• Translational research bridges basic and applied research, applying findings from basic research to practical problems.
• This is often described as "bench to bedside" research.
• Multidisciplinary approaches involve collaborations among different professionals for translational research.
• Basic research -> Translational research -> Applied research
• Research in journals is peer-reviewed.

Research vs. Experience and Scientific Sources

• Research methods provide comparisons; experience lacks comparison groups and confounding variables.
• Research minimizes confounding variables.
• Behavioral research is probabilistic, not definitive.
• Intuition flaws include bias, the availability heuristic (underrepresentation of memorable events), and confirmation bias (favoring supporting information).
• We may be blind to our own biases.
• Scientific, peer-reviewed sources are prioritized over public-access sources.
• Scientific sources include journal articles, chapters in edited books, and full-length books (empirical and review articles, meta-analyses).

Research Claims and Validities

• Research claims involve frequency, association, and causal claims.
• Research validity includes construct, external, statistical, and internal validity.
• Variables are measured or manipulated.
• Operational definitions specify how a concept will be measured.
• Continuous variables have two or more levels; constant variables remain unchanged.
• Research questions are descriptive (measuring), correlational (identifying patterns), or experimental (determining relationships).
• Frequency claims focus on one variable.
• Association claims link two variables.
• Causal claims establish cause-and-effect relationships.

Correlation and Significance

• Correlation coefficients (r) describe the strength and direction of relationships between two variables.
• Positive correlations indicate that high values of one variable are related to high values of the other.
• Negative correlations indicate that high values of one variable are related to low values of the other.
• Zero correlations mean no relationship.
• Statistical validity examines the accuracy and reliability of the study's conclusions.
• Points estimate and confidence intervals (CI) assess the precision of conclusions.

Research Ethics and Validity

• Studies must uphold ethical principles.
• The Belmont Report outlines three ethical principles: respect for persons, beneficence, and justice.
• Research involving animals should adhere to the three Rs (replacement, refinement, reduction).
• Reliability refers to the consistency of a measure, while validity concerns whether the measure assesses the intended concept.

Measuring Variables

• Conceptual variables are abstract constructs being studied.
• Operational variables are concrete measures of concepts.
• Measuring variables can be self-report, observational, or physiological.
• Self-report involves participant responses to questions.
• Observational measurements entail observing behavior or physical traces.
• Physiological measures involve recording biological data.

Categorical vs Quantitative Variables

• Categorical/nominal variables represent distinct categories.
• Quantitative variables represent numerical values.
• Ordinal scales indicate ranked order but not equal intervals.
• Interval scales have equal intervals but not a true zero point.
• Ratio scales have equal intervals and a true zero point.

Reliability and Validity of Measurement Tools

• Reliability assesses consistency; validity ensures the measure assesses the intended concept.
• Test-retest reliability examines consistency across time.
• Interrater reliability measures consistency among observers.
• Internal reliability assesses consistency within a measure.
• Cronbach’s alpha estimates internal consistency.
• Construct validity examines the appropriateness of operational definitions.
  • Face validity concerns if the measure appears to measure what it intends to measure.
  • Content validity assesses if the measure includes all necessary components.
  • Criterion validity examines if the measure correlates with a relevant outcome.
  • Known-groups paradigm distinguishes groups with distinct behaviors.
  • Convergent validity reveals similarity between measures of the same construct.
  • Discriminant/divergent validity finds difference between measures of distinct constructs.

Surveys and Polls

• Construct Validity of surveys relies on effective questioning and responses.
• Open-ended questions allow for varied responses.
• Forced choice questions require selecting between options.
• Likert scales measure agreement/disagreement.
• Semantic differential formats involve rating using anchored adjectives.
• Leading questions can bias responses; double-barreled questions ask multiple things.
• Negatively worded questions lead to confusion.
• Response order and context can influence answers.
• Response sets (acquiescence, fence-sitting, faking) can compromise validity.

Behavioural Observations

• Observational research involves recording behavior.
• Observer bias and effects can distort observations.
• Reactivity occurs when participants change behavior due to being observed.
• Masked/blind design prevents observer bias.

Interviews

• Interviews are a qualitative research method for collecting data.
• Informants are experts on a particular topic and provide insights.
• Interviews on small groups provide specific insights from those with relevant experience.

Sampling

• Populations are the entire groups of interest.
• Samples are subsets of populations.
• Representative samples accurately reflect the population.
• Convenience sampling is easy but biased; self-selection relies on volunteers.
• Probability sampling gives each population member an equal chance of selection.
• Simple random sampling involves randomly selecting participants.