Statistics Flashcards
Standard error of mean
Standard deviation/square root (number of patients)
NNT
1/absolute risk reduction
Variance
SD squared
Study design - outcome measure is relative risk
Cohort Study
Groups selected based on exposure to e.g toxin and followed up to see if they develop the disease
Cohort study
Study design - outcome measure is odds ratio
Case-control
Patients with particular condition identified and matched with controls then look at past exposure to possible causal agent
Case-control
Error: the null hypothesis is rejected when it is true
Type 1, not affected by sample size
Error: the null hypothesis is accepted when it is false
Type 2, affected by sample size
Probability of correctly rejecting the null hypothesis when it is false
Power - increase by increasing sample size
Probability of obtaining a result by chance at least as extreme as the one actually observed assuming null hypothesis is true
P value
Non-parametric, unpaired data
Mann-Whitney U
Non-parametric, paired ‘before’ and ‘after’ on same population
Wilcoxon
Non-parametric, proportions/percentages
Chi-squared
Non-parametric, correlation
Spearman, Kendall