Module 3 observational studies Flashcards
observational studies lack
manipulation
randomization
control
observational designs
cohort: prospective, retrospective
case-control
cross-sectional
cohort studies
group shares common characteristics
longitudinal study
prospective cohort study
moves forward.
population split into 2 groups based upon exposure at that point in time and followed forward to see what diseases occur.
retrospective cohort
looks to past data
population split into 2 groups based upon exposure and what disease states they currently have.
case control study
retrospective
- look at exposure
- start with the individuals current disease status and look back in time to find causality/risk factors
cross-sectional studies
look at one point in time
descriptive: single sample from larger population, no comparison, prevalence studies
analytic: single sample form larger population, compares 2 or more groups in same sample.
observational studies pros and cons
- Pro: many situations not amendable to experimental intervention, large amounts of data looked at in short amount of time, very realistic
- Con: selection bias, association and not causation.
threat to internal validity: history
external event occurred in history at same time as what is being studied, could effect DV
threat to internal validity: maturation
change with passage of time independent of IV
threat to internal validity: selection
self-select groups
threat to internal validity: mortality/attrition
drop-outs
deaths
losses to study groups
-> change in DV
threats to internal validity: testing
pre-test results may affect post-test results
threats to internal validity: instumentation
change implementation mid-study: change a question
calibration is off on tools used
threats to internal validity: design contmaination
participants talk to one another
