University:
University of North Carolina at Charlotte
Course:
HSMT 2101 | Introduction to Health Systems Management
Academic year:

2024
Views:

245

Pages:

3
Author:

customer-8558807

BIOSTATS 711 MASTER CHEAT SHEET WEEK 1 — SUMMARY STATISTICS - Measures: mean, median, mode, variance, SD, IQR, proportions. - Formulas: mean = sum(x)/n, variance = sum((x-mean)^2)/(n-1). - Interpretation: describes center & spread. WEEK 2 — VISUALIZATION - Categorical → bar chart. - Continuous → histogram, boxplot. - Continuous vs continuous → scatterplot. WEEK 3 — PROBABILITY - Addition: P(A or B) = P(A)+P(B)-P(A and B) - Multiplication: P(A and B)=P(A)P(B) if independent. WEEK 4 — DISTRIBUTIONS Normal: z=(x-µ)/σ Binomial: mean=np, var=np(1-p) WEEK 5 — HYPOTHESIS TESTING Steps: State H0/Ha → choose test → compute p-value → compare to alpha → interpret. WEEK 6 — CONFIDENCE INTERVALS CI = estimate ± (critical value × SE) For differences: CI includes 0 → not significant. For ratios (RR/OR): CI includes 1 → not significant. WEEK 7 — ONE-SAMPLE & PAIRED t-TESTS One-sample t = (mean - µ0)/(s/√n) Paired t = mean difference / (SDdiff/√n) WEEK 8 — TWO-SAMPLE t-TEST Use for comparing means between two independent groups. Interpret via mean difference, p-value, CI. WEEK 9 — PROPORTIONS, RISKS, OR/RR Risk = a/(a+b) Odds = risk/(1-risk) RR = Risk1/Risk2 OR = (a/b)/(c/d) WEEK 10 — CHI-SQUARE TEST Categorical vs categorical. Chi-square = Σ (O−E)^2 / E Tests independence. WEEK 11 — CORRELATION & LINEAR REGRESSION Correlation (r): strength of linear relationship. Linear regression: Y = β0 + β1X; β1 = expected change in Y per unit X. WEEK 12 — LOGISTIC REGRESSION Binary outcome. logit(p)=β0+β1X Interpretation: e^(β1)=odds ratio. FLOWCHART FOR SELECTING TEST 1. Outcome continuous: - 1 group vs value → one-sample t - same subjects twice → paired t - 2 independent groups → two-sample t - ≥3 groups → ANOVA 2. Outcome binary: - Compare proportions → 2-proportion test - Categorical vs categorical → chi-square - Predicting binary outcome → logistic regression 3. Two continuous variables: - Association → correlation - Prediction → linear regression