Statistics for Trauma Research Part I: The Bare Essentials: Wax on, wax off I. II. III. IV. V. VI. What is statistics and why do we need it? a. General definition b. Goals of statistics in a clinical setting c. Variability d. Issues/limitations e. Descriptive vs. inferential (analytic) statistics Variables a. Dependent vs. independent b. Variable levels: i. Discrete: nominal, ordinal ii. Continuous: interval, ratio Describing data with numbers a. Central tendency: Mean vs. Median b. Dispersion i. Range ii. Interquartile range iii. Variance and standard deviation Summarizing data visually a. Frequency tables b. Charts, graphs c. Box plots Quantifying extent of disease a. Prevalence b. Incidence i. Cumulative incidence ii. Incidence rate Comparing extent of disease a. Risk difference (aka excess risk, aka attributable risk) i. Attributable fraction b. Relative risk c. Odds ratio Statistics for Trauma Research Part II: Basics of Probability, and types of distributions I. II. III. IV. V. VI. VII. VIII. Basic probability (“likelihood,” “risk”) Conditional probability Sensitivity vs. Specificity Predictive value: positive and negative Independence Binomial distribution Normal distribution Central Limit Theorem Part III: Statistical Inference I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII. Decision matrix P-values Type I, Type II, alpha, Beta Continuous outcome, 1 sample Continuous outcome, 2 samples Categorical outcome, 1 sample Categorical outcome, 2 samples Continuous outcome, 3+ samples: ANOVA Categorical outcome, 3+ samples Statistical vs. clinical significance Confidence intervals P-values vs. confidence intervals Part IV: Nonparametric Tests I. II. III. What are they, and why do we need them? Categorical outcomes Part V: Multivariable Statistics I. II. III. Simple Regression Multiple linear regression Multiple logistic regression Part VI: Sample Size and Power Part VII: Survival Analysis