Stats Flashcards
Relative risk
Experimental event rate/ control event rate
Normal distribution
Mean = median = mode
Standard error of the mean
Measure of spread, gets smaller as sample size increases
SD/ sq root (n)
Absolute risk reduction
Control event rate - experimental event rate
Incidence
Number of new cases per population in given time period
Prevalence
Total number of cases per population at a particular point in time
Sensitivity
Ie picks up all cases
Negative result can rule out disease
Power
Increased by sample size
= 1 - probability of type 2 error
Type 1 error
Null hypothesis rejected when true
Type 2 error
Null hypothesis accepted when it’s false
Specificity
Ie recognises all healthy people as healthy
Number of true -ve / (true -ve + false positives)
Cohort study
Selected according to exposure to particular agent and followed up to see how many develop outcome
Case control
Rare conditions
Patients with certain condition identified and matched with controls
Cross sectional study
Snapshot
Weak evidence of cause and effect
Observational study