surveys Flashcards
(68 cards)
cross sectional study
everything is measured at the same time
all the data is available - exposures and outcomes.
Attributable Risk can be calculated by:
Incidence exposed - Incidence unexposed
Attributable Risk Fraction can be calculated by:
Incidence exposed - Incidence unexposed/Incidence exposed
How can a confounding factor be dealt with in design stage of study(3)
Randomisation
Stratification
Restriction
Incidence rate ratio be calculated
incidence rate in exposed/incidence rate in unexposed
How can instrument bias be avoided
Assess how well instrument performs over time in different settings with different groups
How can instrument bias be reduced
Good design of questionaires
calibrate and maintain equipment
How can interviewer bias be reduced
Training
closed questions
have a criteria for outcome of the assessment
How can non-response bias be reduced
Offer incentives
Explain reasoning
Reminders
relative risk
cumulative incidence in exposed/ cumulative incidence in unexposed
How can sampling bias be reduced
Random sampling
How is SMR (standardised mortality ratio) expressed
As a percentage
selection bias
groups are not similar at baseline because of nonrandom assignment.
validity
ability of a test to measure what it is intended to measure.
precision
analogous to reliability and a measure of a test’s ability to replicate the same results.
observational study
investigator observes the subjects without intervention
Attributable risk
increase in disease incidence associated with an exposure.
Accuracy
analogous to validity and a measure of a test’s ability to obtain repeatedly “true” results.
Why is a sampling frame used
It is the source material or device from which a sample is drawn
What quality control needs to occur in study
Standardise instruments, regular calibration
training for interviewers
structured questionnaires
What is the purpose of a survey
Assess prevalance of disease
measure risk and protective factors
measure outcomes
What is the definition of confounder?
Third variable associated with both exposure and outcome
What is systematic sampling
Every Nth member selected
What is stratified sampling
Population divided into groups and sample taken from each group using simple random sampling