2.3 Research Methodology

McMaster University **We aren't endorsed by this school
Oct 16, 2023
Uploaded by GeneralSnow10855 on coursehero.com
RESEARCH IS COMPLICATED Click bait and social media expediates that spread of false or sensationalized ideas An understanding of how research works is important Understanding biases and know where media 'over extrapolates findings' EXPERIMENTAL DESIGNS Experiment - a test under controlled conditions Random assignments to groups is critical (RCTs) Placebo Effect - the tendency for people to positively evaluate any treatment, even though there may no physiological effects attributable to the treatment Dose - Response Curve = plots the symptoms against amount of medication Determines amount of toxicity in substances & symptoms Design Issues for Human Aging Research Length of human aging Cant follow people forever since we also age - Animal research does not always translate to humans Biology is similar but not identical - Use human cells (in-vitro), rodents, round worms, yeast, flies - Use of surrogate outcome variables Correlation is not always perfect though - Ex. Measure blood cholesterol when interested in heart disease - Ethics of treatment Can't try something that might actually reduce lifespan - Cohort effects 2.3 Research Methodology Thursday, September 14, 2023 11:59 AM
Quasi-Experimental Designs Resembles experimental research, but not truly; independent variables can be manipulated but participants are not randomly assigned Used when RCT is not practical or ethical Compare differences between groups Statistically control some aspects (who we let in) - Causal language should be avoided Associations only - Case control design for rare phenomenon Not randomized - May overestimate the effect size - Ex. Blue zone is the case, then find control in a similar area - Case Studies & Human Disease States In-depth and detailed examination of a particular individual, group, or event in real life settings Often exploratory Not very generalizable - Human disease states may provide insights into aging Not independent of confounding factors - Correlational Designs Association among groups of variables Strength of relationship is looked at Causal Directionality = cannot be certain which variable cause the others Cross-Sectional vs Longitudinal Designs Age Differences --> use cross sectional Compare different age groups at one point in time - Very susceptible to cohort effects - Age-Related Changes --> use longitudinal designs
Follow ppl over time - Disadvantages of Longitudinal Studies: Fewer respondents at each assessment Bc death, dropout, moving - Cohort effects Period effect = a change that occurs at a particular time but affect all age groups and all cohorts uniformly Meta-Analyses Summarizing the literature and developing hypothesis to explain contradictory findings Use forest plot Summarizes study findings, directions, overall - STATISTICAL CONSIDERATIONS IN AGING RESEARCH Means are different - does not mean that applies to all Direct Effect: drinking --> health Indirect Effect: drinking --> increased PA --> health Mediated Effects: social ties --> drinking AND social ties --> health
Page1of 3
Uploaded by GeneralSnow10855 on coursehero.com