Concepts of research Flashcards
Interventional studies
Study designs considered experimental
Researcher-forced group allocation
Randomization commonly used
Investigator assigns exposures
Observational studies
Designs considered natural
Most observational studies not able to prove causation
An observational study with no comparison group would be called
Descriptive study
Study design selection based on
Perspective of research question/hypothesis
Desire of researcher to force group allocation
Ethics of methodology
Efficiency and practicality
Costs
Validity of acquired information
External validity/generalizability
Study population selection based on
Hypothesis
Population of interest
Inclusion & exclusion selection criteria (interventional)
Case&control group or exposed/non exposed group selection criteria (observational)
Null hypothesis
A research perspective which states there will be no true difference between the groups being compared
Type 1 error
False positive
Type 2 error
False negative
Probability samples
Every element in the population has a known (non-zero) probability of being included in a sample
Simple random sampling
Assign random numbers, then take randomly selected numbers to get sample size or
Assign random numbers, sequentially list numbers and take sample size from top or bottom of list
Systematic random sampling
Assign random numbers, randomly sort numbers, select highest or lowest number and then take every Nth number from there
Stratified simple random sampling
Stratify sampling frame by desired characteristics (e.g., gender) then use simple random sampling to select sample size
Stratified disproportionate random sampling
Disproportionately use stratified simple random sampling when baseline population is not at the desired proportional percentages to the reference population
Stratified sample is weighted to return sample population back to baseline
Multi-state random sampling
Uses simple random sampling at multiple-stages toward patient selection Regions/counties Zip codes Household Individuals
Cluster multi-stage random sampling
Same as multi-stage but all elements clustered together (at any stage) are selected for inclusion
All clinics in a zip code
All households in a community