Cross sectional studies Flashcards
We want sample to be…
Precise
Unbiased
Representative
Concept of ecological studies
Snapshot in time and estimate at group level
No reflection of time lag of diseases etc
How to conduct an ecological study?
Identify groups
Define characteristics - exposure and outcome
Decide if analysis uses categorical data or continuous data
Gather data on group level stats
Discrete/qualitive data types
Nominal (categories eg ethnicity)
Ordinal (ordered categories eg stage of cancer)
Continuous data types
Interval (meaningful differences eg calendar year)
Ratio (absolute zero eg weight)
Specific issues for ecological studies
Definition of characteristics
Measurement variation
Confounding (ecological fallacy)
Chance/random error
What is ecological fallacy?
Falsely inferring individual level association from group level association
Concept of cross sectional studies
Snapshot in time
Individual level analysis
What are cross sectional studies usually used for?
Find out prevalence
What is prevalence?
Proportion of people with a disease at a given point in time (new and old cases)
Example of cross sectional study
BMI of individuals and how much fast food they eat weekly
Example of ecological study
Daily meat consumption (countries) and Colon cancer incidence
Issues with cross sectional study
Timing - does timing of exposure relate well to the current BMI (only snapshot)
Bias - is self reported data true? missclassification?
Confounding
Random variation
What does it mean if null hypothesis value in regards to confidence interval?
If within interval - results are consistent with hypothesis
If outside interval - results are not consistent, p is less than 0.05, evidence to reject null hypothesis
What would null hypothesis value be?
If ratio: 1
If a difference: 0