Qualitative Research Overview

Qualitative Research

• Involves collecting and analyzing non-numeric data (e.g., language, text)
• Applied across disciplines like anthropology, psychology, sociology, and kinesiology
• Also known as ethnographic, grounded theory, phenomenological, etc.
• The research process includes defining the research problem, formulating research questions and a theoretical framework, and collecting data
• Data collection methods include interviews, focus groups, and observations
• Trustworthiness is crucial for ensuring the applicability, consistency, neutrality of data
• Trustworthiness components include credibility, transferability, dependability, and confirmability
• Prolonged engagement, audit trails, thick descriptions help with data quality
• Understanding researcher bias and minimizing bias is crucial for ensuring trustworthiness
• Triangulation (using multiple sources for supporting conclusions) and member checking (validating findings with participants) are key techniques for ensuring data validity

Experimental & Quasi-Experimental Research

• Used to establish cause-and-effect relationships
• A good theoretical framework, experimental design, appropriate variables and statistical analyses are crucial
• Criteria for establishing causality includes: the cause must precede the effect, cause and effect must correlate, and no other variable should explain the correlation
• Types of validity include internal validity (ensures the treatment cause observed changes) and external validity (generalizability to other contexts)
• Threats to internal validity include history, maturation, testing effects, selection biases
• Threats to external validity include interaction of testing with treatment, biases in participant selection
• Experimental designs include pre-experimental, experimental, and quasi-experimental designs

Epidemiology

• Involves the study of the distribution and determinants of health-related states or events in specified populations, and the application of this study to the control of health problems
• Observational research examines natural differences without intervention, while experimental research tests the effects of treatments (ethically limited)
• Key epidemiological concepts include distribution (frequency, patterns) and determinants (characteristics affecting health)
• Common study designs include cross-sectional studies, cohort studies, and case-control studies

Correlation & Regression

• Correlation examines relationships between two or more variables
• Types of correlation include positive (direct relationship) and negative (inverse relationship)
• Large samples improve the detection of significant relationships
• Regression predicts outcomes based on one or more predictor variables
• Simple regression predicts outcomes based on a single predictor variable, while multiple regression uses several predictors for a more complex outcome
• Logistic regression predicts binary outcomes (e.g., yes/no)

Research Synthesis and Meta-Analysis

• A research synthesis defines a clear purpose before starting a literature review, while a literature review is often exploratory, focused on identifying a research purpose
• Meta-analysis involves analyzing data from multiple studies using a standard metric
• Key steps in a meta-analysis include identifying the problem, searching for relevant literature, reviewing studies, evaluating and coding data, calculating effect sizes, performing statistical analysis, and reporting findings
• Meta-analysis can be used to answer research questions, synthesize research findings, and draw conclusions from multiple studies
• Systematic reviews are used to summarize research on a specific topic without the quantitative focus of meta-analysis
• Key steps in a systematic review include forming a team, registering the review, screening studies for eligibility, and including relevant studies in the review
• Flowcharts are used to visually track the steps in a systematic review, from identification to inclusion of studies
• Considerations in meta-analysis include effect size coding, homogeneity testing, and bias control
• Systematic reviews are essential for evidence-based practice because they provide a comprehensive summary of the available research on a specific topic, promoting informed decision-making in various fields