Skip to main content

Misc Final Notes

TODO​

  • How do you account for variability in the data?
  • Bias and types of bias.
  • Important differences in qualitative methods
    • Observation: You see but you cannot ask why. You don’t get motivations.
    • Interviews: Ask for motivations, attitudes, perspectives. Hard to interview about implicit practices (how to ride a bicycle, what do you do before you go to bed, ask nurses how they care for a patient).
    • Hybrids

Concept, Construct, Variable​

  • Concept: “Health”
  • Construct: “Cardiovascular Fitness”
  • Variable: VO2\text{VO}_2 max or HR or Resting HR or…

Efficacy and Effectiveness​

See this. It’s generally about intervention outcomes.

Belmont Report​

  • Respect for persons
    Individuals as autonomous agents, informed consent, disclosure, confidentiality, additional protections for diminished autonomy, avoid coercion/excessive compensation
  • Beneficence
    Do no harm, maximize benefits, scientifically sound design, acceptable risk-benefit ratio, minimize physical and psychosocial harm
  • Justice
    Fair distribution of benefits and burdens, vulnerable populations, fair distribution of research

Falsifiability​

You cannot confirm theory with empirical observation. You can only refute it. Thank Mister Popper. This is the basis for why the Null Hypothesis is what it is in hypothesis testing.

Black swan example: It takes one black swan to disprove “All Swans are White.”

Logical Positivism/Empiricism​

Logical positivism’s central thesis was the verification principle, also known as the “verifiability criterion of meaning”, according to which a statement is cognitively meaningful only if it can be verified through empirical observation or if it is a tautology (true by virtue of its own meaning or its own logical form).

Translational Research (R18 Grant)​

Translational research accelerates the movement of basic scientific discoveries from the laboratory and preclinical studies into clinical applications, diagnostics, and treatments to improve human health.

Synthesis​

  • Metasummary (Qualitative)
    Sum up findings in some area.
  • Metasynthesis (Qualitative)
    Use studies snad metasummaries to synthesize
  • Metaanalysis
    Compute effect size. “The pooled odds ratio is 1.34 (95% CI 1.12–1.60)”
  • Systematic Review
    ”Across 47 studies of varying quality, the evidence suggests X, with these caveats.” Systematic Review is the methodology for identifying and evaluating the evidence. Meta-analysis is the statistical method for combining it.
  • Integrative Review
    ”What is known about how adults with type 2 diabetes use mobile health apps to self-manage their condition?” Note that Systematic Review pins itself typically to one study type but IR can span several.
  • Scoping Review
    What a PhD student would do. What kinds of studies exist, what populations, what outcomes are measured. No synthesis.
  • Narrative Review
    integrative review but without the methodological rigor: no protocol, no systematic search, no formal quality appraisal, just some expertbro’s curated tour of the literature.

Here’s an intro paper to Systematic Review and Metaanalysis

Validity, Reliability, Generalizability​

  • Validity: How true are your results? Did you measure what you wanted to measure?
  • Reliability: Are your results consistent over time? Can they be reproduced?
  • And: Can your results can be generalized to other populations that were not part of your study?