General Concepts of Research Flashcards
types of quantitative study designs
interventional and observation
interventional study design
considered “experimental”
forced allocation to study groups
investigator selects interventions (exposure)
observational study design
considered “natural”
no forced allocation to study groups
researchers “observe” subject-elements occurring naturally or selected by individual (naturally or freely)
useful for unethical study designs using forced interventions
no designed to prove causation
Interventional and Observation study design strength of evidence
Interventional Observational
- Phase 0 -Case Reports/Series
- Phase 1 -Ecological (not important)
- Phase 2 -Cross-Sectional
- Phase 3 -Case-Control
- Phase 4 -Cohort
*Increasing strength downward
Research Evidence Pyramid
Systematic Reviews/Meta-Analyses Interventional (Explanatory)/Pragmatic (Exploratory) Cohort Case-Control Ecological (not important) Case Series Case Reports Animal Research In vitro (test-tube) research
Research question
Purpose statement; formed as a question
Helps frame study intent and can direct researcher to selecting and developing an effective study design and methods to answer question
Study design and methods selection is based on what?
- perspective of research question
- ability/desire of researcher to force group allocation and intervention (randomization)
- ethics of methodology selected
- efficiency and practicality
- costs
- Internal validity (validity of acquired information)
- external validity
population
all individuals making up a common group; from which a sample can be obtained
sample
a subset or portion of the full and complete population
interventional study population is based on what?
research hypothesis
population of interest
inclusion and exclusion criteria (impacts external validity)
observational study population is based on what?
research hypothesis
population of interest
Case and Control group OR Exposed and Non-Exposed group selection criteria (impacts external validity)
Null hypothesis (Ho)
research perspective which states there will be no true difference between the groups being compared
Odds ratio is 1 between groups
type I error
false positive
rejecting a true null hypothesis
type II error
accepting (fail to reject) a false null hypothesis
statistical-perspectives used in deciding to reject or not reject null hypothesis
- superiority: is the drug better than placebo? OR better than another drug?
- noninferiority: drug is not worse than the placebo
- equivalency: both drug and the placebo are the same
prospective study
Don’t know outcome yet
groups are followed into future to assess for outcomes of interest and then groups are compared
used by all interventional studies and some observational studies
retrospective study
outcome known at start of study
outcomes of interest have already occurred; researchers have goal of assessing past history of some activity (exposure/treatment)
ambidirectional
uses retrospective design to assess past differences in occurrences of events (exposure or outcome), but also adds future data collected on additional outcomes prospectively from start of study
probability samples
every element in population has a known (non-zero) probability of being included in sample
most common
simple random sampling
type of probability sampling
assign random numbers, then take randomly-selected numbers to get desired sample size
OR
assign random numbers, then sequentially-list numbers and take desired sample size from top (or bottom) of listed numbers
stratified simple random sampling
type of probability sampling
stratify sampling frame by desired characteristic (e.g. gender), then use simple random sampling to select desired sample size
multi-stage random sampling
type of probability sampling
uses simple random sampling at multiple-stages towards patient selection
e.g. –> counties –> zip codes –> households –> individual
non-probability sampling scheme
not a known probability
can produce selection bias
quasi-systematic or convenience samples
not really completely random or fully probabilistic
decide on what fraction of population is to be sampled and how they will be sampled
e.g. all last name M-Z; all persons attending clinic every M/W/F for 6 months