Title  Contributor/Contact  Keywords 

Alabama  Simple Descriptive Statistics  George Howard, DrPH  Infant mortality rates Descriptive statistics Analysis of variance Histograms  intervals and scales Measures of central tendency: mean, median Measures of dispersion: range, percentile, standard deviation, variance Outliers Box and whisker plots Skew 
Alabama  Introduction to Random Variables  George Howard, DrPH  Dichotomous Outcome Continuous outcome Tossing coins Logic and exclusion approach Mathematical approach Generalizations Distributions Binomial distribution The Bell Shaped Curve 
Alabama  Multivariate Statistics  George Howard, DrPH  Multivariate regression (joint effect) Example 1: Age, SBP IMT, univariate, multivariable Example 2: SBP and Age  confounding Example 3: Age and SBP related to IMT  confounding Example 4: Age and SBP inconsistently related to IMT  effect modification Example 5: Both Age and SBP related to IMT  joint effects 
Alabama  Sampling Distributions 
George Howard, DrPH  Estimation Universe Parameters How to estimate parameters Standard deviation Reliability Standard error Confidence limits on the mean Interpretation of the confidence limits on the mean REGARDS Study data  mean Binomial distribution Standard error of a binomial 
Alabama  Vocabulary of Uncertainty  George Howard, DrPH  Statistics = common sense; Inference; Sample; Estimation; Strength of Association; Hypothesis Testing; PValue; PostHoc Relationships; Error; PValue vs Power; Confounder; Fixing Confounders  Matching; Fixing Confounders  Adjust; Confounders  What can go wrong?; Confounders in RCT; Selection of Statistical Tools; Example Stat Tool  BP and Gender; Example Stat Tool  BP and Age; Types of Statistical Tests 
Columbia  Statistics for the Basic Sciences  Ivey Chen  Assess data sources probability and inference deductive reasoning systematic error measurement error hypothesis testing type I and type II errors confidence bounds 
Duke  EvidenceBased Medicine Education Series  Karen Pieper, MS  Evidencebase medicine Observational studies Randomized clinical trials Statistical issues in design of clinical trials Randomized studies Nonrandomized studies Randomization process Endpoints Multiple comparisons Data validation Modeling Data safety monitoring boards 
Duke  Lesson 1: EvidenceBased Medicine: The Basic Concepts  Karen Pieper, MS, Robert A. Harrington, MD  What is EBM? Why does EBM matter? Why is EBM important to drug development? Observational studies Randomized clinical trials Types of trials Concepts (randomization, sample sizes, etc.) Regulatory issues on clinical trials 
Duke  Lesson 2: Statistical Issues in the Design of a Trial, Part 1  Karen Pieper, MS  Experimental design Randomized Nonrandomized studies Simple randomization Blocked randomization Stratification Randomization process Group allocation design Factorial design Crossover design Blinding Phases of a study 
Duke  Lesson 3: Statistical Issues in the Design of a Trial, Part 2  Karen Pieper, MS  Endpoints "Hard" versus "soft" endpoints Adjudication of endpoints Timing of endpoints Surrogate endpoints Single versus composite endpoints Intention to treat Primary hypotheses Secondary hypotheses Multiple comparisons Calculating sample size 
Duke  Lesson 4: Interpretation of Medical Literature Statistics and Statistical Terms  Karen Pieper, MS  
Duke  Lesson 5: Statistical Methods for Analyses  Karen Pieper, MS  Data validation Patients who do not meet eligibility criteria Incorrect data Missing data Protocol violations Modeling Cindex Binary shortterm outcomes Binary longterm outcomes Continuous outcomes Parametric versus nonparametric tests Independent versus dependent variables 
Duke  Lesson 6: The Role of Independent Data Monitoring  Karen Pieper, MS  Data monitoring Stopping rules Data safety monitoring boards 
Indiana  Analysis of Categorical Data  Siu L. Hui, PhD  Inference Estimation Hypothesis testing Categorical variables Binary variables Examples of onesample problems Point estimate Confidence interval Example: Phase 2 clinical trial Example: Aminocentesis and pregnancy loss Prospective randomized controlled trial Retrospective case control study Crosssectional study Example: Randomized study to reduce antibiotic use ChiSquare Tests: Large Sample (RxC) ChiSquare Tests: Fisher's exact test ChiSquare Tests: McNemar's test 
Indiana  Basics of Clinical Data Management  Tim Breen, PhD, MS, CCDM  informatics process data quality research proposal protocol ICH GCP stages of clinical trial data management plan data collection tools CRF content CRF layout database validation quality assurance quality control 
Indiana  Comparison of Means  Susan M. Perkins, PhD  Choice of Test: Study Design, Distribution, Number of Groups Student's TTest: Paired Design Student's TTest for Independent Groups Assumptions: TTests TwoGroups  NonNormal Data: Paired or Independent Sample size calculations Negative results Choice of sample size depends on: Question, variability, and analysis Analysis of Variance (ANOVA) Assumptions for ANOVA: Normality, equal variances, independent observations Example of ANOVA  CD4 Counts and PTSD in HIV Patients Multiple Comparison Techniques 
Indiana  Correlation and Simple Linear Regression  Chandon Saha, PhD  Correlation analysis Pearson correlation Spearman rank correlation Simple linear regression Scatterplots for correlation Example of correlation for height and weight Interpretation of the size of r  explain variability When not to use r Assumptions about the random error Fitting the linear model Principle of Least Squares Extrapolation Causality Simple linear regression Residuals Normality assumption 
Indiana  Design of Genetic Studies  Dan Koller, PhD  Molecular markers Microsatellite markers Single Nucleotide Polymorphisms (SNPs) Single gene (Mendelian) disorders Genetically complex disorders Twin studies Familial aggregation Meiosis and linkage Genome screen approach LOD score Potential candidate approach Linkage vs association Linkage Disequilibrium Studies Association Studies Population based approach  case control design Family based approach  Transmission Disequilibrium Test (TDT) 
Indiana  Evaluation of Diagnostic Tests  Siu L. Hui, PhD  Screening test Diagnostic test Accuracy of a diagnostic test Measure of agreement of two tests Example: Staging of Prostate Cancer with MRI True disease status Test result Sensitivity Specificity False positive rate Predictive value positive Predictive value negative Example: Screening Test for a Rare Disease Receiver Operating Characteristic (ROC) Curve Continuous test results Example: Blood Test for a Disease Reliability Kappa Intraclass Correlation (ICC) Example: Biphasic Radiography vs Fiberoptic Endoscopy in Gastric Ulcers 
Indiana  Hypothesis Testing and Confidence Interval Estimation  Susan M. Perkins, PhD  Statistical inference Example of children with new onset seizures and neuropsychological functioning Example of women of child bearing age and caffeine consumption Hypothesis testing Null and alternative hypothesis Test statistic Rejection Region Type I Error Type II Error Sample size and power One sample test of the mean Central Limit Theorem Steps in hypothesis testing Significance level Rejection region Relationship of clinical and statistical significance Confidence interval 
Indiana  Introduction to Longitudinal Studies  Zhangsheng Yu  Longitudinal studies Cohort study Intraindividual correlation Interindividual variation Changes over time Crosssectional studies Reading ability example Exploratory analysis Confirmatory analysis Ad hoc analysis Random effect models Marginal models Markov transition models Graphical explanation of the random effects model 
Indiana  Introduction to Survival Analysis  Zhangsheng Yu  Survival analysis Time to event analysis Logistic regression Censored observations Staggered entry of patients Right censoring Left censoring Interval censoring Example of children with acute leukemia Mean followup Mean survival Median survival KaplanMeier curves Crook study of prostate cancer Cox proportional hazards model 
Indiana  Issues in Clinical Trials  Chandon Saha, PhD  Definition of a clinical trial Phase I clinical trial Phase II clinical trial Phase III clinical trial Phase IV clinical trial Dosefinding trial Feasibility Treatmentefficacy Comparative efficacy Postmarketing observational studies Three basic ethical principles: Autonomy Beneficence Justice Clinical trial design Features of welldesigned clinical trials Phase I design Comparative Study (Phase III) design: Two groups parallel study Crossover trials Randomization Stratification Randomization using permuted blocks Sample size determination Power Monitoring a clinical trial Groupsequential methods: Pocock method O'brienFleming design Descriptive statistics Statistical inference Components of a study report Deficiencies in trial reporting Deficiencies in reporting of results Deficiencies of editorial standards 
Indiana  Multiple Linear Regression and Logistic Regression  Patrick Monahan, PhD  Simple linear regression Independent variable Dependent variable Predictor or explanatory variable Outcome or response variable Multiple linear regression Assumptions of linear regression Assumptions of multiple linear regression Example  multiple linear regression  Perceived barriers to mammography screening F test for significance Effect sizes Partial ttest Individual covariates Semipartial correlation coefficient Graphical model checking Nonnormality Nonlinearity Nonconstant variance Residuals Outliers Logistic regression Coding Wald chisquare tests Adjust odds ratio Likelihood ratio chisquare tests Model selection  explanatory Model selection  prediction 
Indiana  Study Design  Patrick Monahan, PhD  Observation Informal experimentation Formal experimentation: Evidence based medicine and Causal inference Elements of a study Question and Testable Hypothesis Outcome variable and its measurement Experimental study design Population Protocol Analyses Conclusion should answer question Epidemiologic study designs Case Series Assess association between risk factors and disease Ecologic CrossSectional CaseControl Prospective and Historical Perspective Measures of Disease Frequency Prevalence Cumulative Incidence Incidence Rate Risk vs Rate Ecologic Studies Example of Ecologic Study  Breast Cancer Advantages and Disadvantages of Ecologic Studies CrossSectional Studies Example of CrossSectional Study  NHANES Advantages and Disadvantages of CrossSectional Studies CaseControl Studies Measures of Association  CaseControl: Odds Ratio 2x2 Table Example of CaseControl Study  Smoking Example of CaseControl Study  Febrile Seizures Advantages and Disadvantages of CaseControl Studies Prospective Studies Example of Prospective Study  British Male Doctors and Smoking Measure of Association  Prospective Studies: Relative Risk, Attributable Risk, and Population Attributable Risk Advantages and Disadvantages of Prospective Studies Historical Prospective Studies Example of Historical Prospective Study  Allegheny Steelworkers Advantages and Disadvantages of Historical Prospective Studies Case Control vs Cohort Studies Outcome Studies Common Outcome Measures: Morbidity, mortality, disease severity (PASI, NYHA), health status (SF36, SIP), Quality of Life (QWB) Advantages and Disadvantages of Health Care Delivery Studies 
Lecture 1: Introduction to Stata  Introduction to Stata Software packages used in clinical research Why use Stata? Why not SAS or R? Stata basic functionality Menu vs command line Describing your data Descriptive statistics  continuous data Graphical exploration  continuous data Descriptive statistics  categorical data Analytical statistics  2 categorical variables Analytical statistics  1 categorical, 1 continuous Analytical statistics  2 continuous 

Lecture 2: Do files, log files, and workflow in Stata  Do files Comments Advantages of Do files Log files Advantages of Log files Using Do and Log files together 

Lecture 3: Generating new variables and manipulating data with STATA  Importing into Stata Exploring your data Cleaning your data Making a new variable Manipulating values of a variable Getting rid of variables Labeling variables Missing values 

Lecture 4: Using Excel  
Lecture 5: Basic epidemiologic analysis with Stata  Statistical analysis Epidemiologic analysis Confounding Interaction Causal diagrams 2x2 or Contingency tables Effect modification Multivariable adjustment 

Lecture 6  Part 1: Making a figure with Stata or Excel  Figures Photographs Diagrams Pie charts Scatter plots Bar graphs Line Graphs 

Lecture 62: Mostly Dates\x85and a few other useful STATA commands  Date conversion Date formatted Compare dates Literal dates Programming loops Merging datasets 

Lecture 7: Organizing a project, making a table  Organizing Stata files Table components Stata to Word Stata to Excel 

Lesson 2: Statistical Issues in the Design of a Trial, Part 1  Experimental design Randomized Nonrandomized studies Simple randomization Blocked randomization Stratification Randomization process Group allocation design Factorial design Crossover design Blinding Phases of a study 

MCW  ANOVA: Comparing More Than Two  Sergey Tarima, PhD  ANOVA More than two treatments 
MCW  All Biostatistics YouTube Videos  MCW Biostatistics  
MCW  Analyzing Discrete Data  Kwang Woo Ahn, PhD  Discrete data 
MCW  Choosing Statistical Software  Brent Logan, PhD  Choosing Statistical Sofware SAS SPSS STATA Cost of software packages 
MCW  Concepts in Biostatistics  Prakash Laud, PhD  Concepts Data Decisions Hypotheses formulation Confidence interval 
MCW  Designing Clinical Trials  Brent R. Logan, PhD  Designing Clinical trials Phase I Trials Phase II Trials Phase III Trials 
MCW  Discrete Data Analysis  Tao Wang, PhD  
MCW  Equivalence and NonInferiority Testing  Gisela Tunes da Silva, PhD  Equivalence NonInferiority Testing Methods Same Not worse 
MCW  Introduction to Survival Analysis  John Klein, PhD  Survival Analysis Survival Data Univariate Statistics KaplanMeier Estimator Competeing Risks 
MCW  Logistic regression  Sergey Tarima, PhD  Logistic regression 
MCW  Matched Studies in Medical Research  John P. Klein, PhD  Matched Studies in Medical Research Matched Pairs Design Examples of Biological Matching John Klein Medical College of Wisconsin 
MCW  Multiple Comparisons  Brent R. Logan, PhD  Multiple comparisons 
MCW  Paired Data  Ruta Bajorunaite, PhD  Paired data 
MCW  Paired Data Analysis  Jennifer LeRademacher, PhD  paired data analysis independent vs. paired data paired ttest sign test Wilcoxon Sign rnk test McNemar's test 
MCW  Reading Medical Literature  Ruta Brazauskas, PhD  Reading Medical Literature Parts of a paper Medical College of Wisconsin Ruta Brazauskas 
MCW  Simple Linear Regression  MeiJie Zhang, PhD  Simple linear regression Equation of a line Fitting line to data Strength of Association Assumptions of Linear Regression 
MCW  Simple Statistics in Excel  Aniko Szabo, PhD  Descriptives Pivot Table Correlation Ttest Paired Ttest Regression ANOVA 
MCW  Statistical Considerations in Grant Writing  Aniko Szabo, PhD  Statistical Considerations Grant Writing Mistakes and pitfalls in grant writing statistical issues in grant writing 
MCW  Statistics, Probability & Diagnostic Medicine  Jennifer LeRademacher, PhD  Diagnostic Accuracy Statistics for Qualitative Tests Sensitivity Specificity Probability 
MCW  Uses and Abuses of NonParametric Statistics  John Klein, PhD  Nonparametric statistics Medical Research Nonparametric tests Medical College of Wisconsin John Klein 
MCW  WebBased Sample Size Calculation  Kwang Woo Ahn, PhD  Sample size calculation Basic power calculations Sample size analysis 
MCW  Writing a Protocol  John Klein, PhD  Protocol 
MCW Basic Concepts of Bayesian Statistics  Prakash Laud, PhD  Bayesian statistics calibration of probability Bayes Theorem Bayesian methods adaptive clinical trials 
MCW Common Errors in Linear Regression  Kwang Woo Ahn, PhD  
MCW Getting Help for Your Biostatistics Questions and Database Basics  Dan Eastwood, MS  MCW statistical help Medical College of Wisconsin Milwaukee Dan Eastwood database basics biostatistics questions organize data types of data spreadsheet vs database data management 
MCW Propensity Scores  MeiJie Zhang, PhD  propensity score propensity score method propensity score estimation method propensity score matching estimating treatment effect using propensity including matched pair analysis and regression adjustment and stratification 
Mayo  Statistics in Clinical Research  Felicity Enders, PhD  
NIH  A Conceptual Approach to Survival Analysis  Laura Lee Johnson, PhD  Survival or time to event Median survival Outcome variable Event time Time origin Time scale Sample size Truncation Censoring Right censoring: withdrawals Independent censoring Covariates Survival Hazard fundtions Competing risks Independence assumption Kaplan Meier Inference Log rank Stratified log rank Cox Proportional Hazards Model TimeDependent Survival Curves 
NIH  Design of Epidemiological Studies  Laura Lee Johnson, PhD  Questions Hypothesis Experimental design Samples Data Analyses Conclusions Epidemiology Relative risk Odds ration Risk difference Doseresponse relationships Temporal sequence Consistency of the association (internal "validity") Replication of results (external validity) Biological plausibility Experimental evidence Variable Covariate Coefficient Association Confounders Effect modifiers Example: Coffee and pancreatic cancer Randomization Prevalence Incidence Sensitivity Specificity Selection bias Interviewer bias Randomized vs nonrandomized study Blinded/Masked or not Treatment vs observational study Prospective vs retrospective study Longitudinal vs crosssectional study Case reports Mean Standard deviation Proportions Confidence limits or intervals Casecontrol study Cohort study 
NIH  Issues in Randomization  Laura Lee Johnson, PhD  Randomized Study Design Sample size Mean Median Variance Standard deviation Standard error Odds ratio Sampling distribution Quasi experimental studies Blocked randomization Stratified randomization Cluster randomization Adaptive randomization Intent to treat Parallel group Sequential trials Group sequential trials Crossover Factorial designs Adaptive designs 
NIH  Principles of Hypothesis Testing for Public Health  Laura Lee Johnson, PhD  Estimation Statistical Inference Hypothesis testing Distributions Sides and tails Clinically important difference Point estimation Interval estimation Scatter plots Box plots Histograms Bar plots Null hypothesis Alternative hypothesis Twosided test Onesided test Type I error Type II error Significance level Power Confidence interval Linear regression Repeated measures Positive predictive value Negative predictive value Sensitivity Specificity 
Rochester  Basic Statistics: MBI540 (for Virologists)  Sally Thurston, PhD  Experimental Design Mean Standard Deviation Standard Error M&M Data 2Sample Ttest Importance of Plotting the Data 1Way ANOVA 
UCI  A Brief Introduction to R  Babak Shahbaba (babaks@uci.edu)  R Objects Vectors Matrices Data frames Loops 
UCLA  Online Biostatistics Courses and Seminars  UCLA Biostatistics Consulting  (See individual classes) 
UCSF  Case Control Study Design  Jeff Martin, MD  
UCSF  Clinical Trial Design Video  Deborah Grady, MD  
UCSF  Cohort Study Design  Jeff Martin, MD  Cohort study design Threats to validity Framingham Study HCV/HIV Study 
UCSF  Common Biostatistical Problems  PeterBacchetti  Pvalues for establishing negative conclusions Misleading and vague phrasing Speculation about low power Exclusive reliance on intenttotreat analysis Reliance on omnibus tests and Overuse of multiple comparisons adjustments Entangled outcomes and predictors 
UCSF  CrossSectional Study Design  Jeff Martin, MD  Crosssectional study Point Prevalence Period Prevalence 
UCSF  Data Exploration  David Glidden, PhD  Multiple predictor regression Date Types: Numerical (Continuous and Discrete), Categorical (Ordinal and Nominal) Hierarchy of data types Outliers Boxplots Histograms 
UCSF  Disease AssociationRisk Ratio  Jeff Martin, MD  Disease association Risk ratio Odds ratio Log Rank Test Wilcoxon 
UCSF  Intro Cohort Study Design  Jeff Martin, MD  Cohort study design Human subjects studies Study base 
UCSF  MultiPredictor Linear Model  David Glidden, PhD  Linear Regression Predictors Normality Regression dilution bias Regression coefficients Negative mediation Logtransformed predictors Confidence intervals 
UCSF  Simple Linear Regression  David Glidden, PhD  Scatterplot Mean 95% CI Linear regression Least squares Outlier Bias Confounding 
UCSF  Stata Course  Mark Pletcher, MD, MPH  (See individual lectures.) 
Vanderbilt  Biostatistics II Lecture Notes for 2010  William C. Dupont, PhD  Simple Linear Regression Multiple Linear Regression Simple Logistic Regression Multiple Logistic Regression Survival Analysis Hazard Regression Simple Poisson Regression Multiple Poisson Regression Analysis of Variance Mixed Effect Analysis of Variance 
Vanderbilt  Teaching Methods  Biostatistics  Vanderbilt Biostatistics  (See individual courses) 
www.SampleSizeShop.org  Keith Muller/Deb Glueck  Sample size, mixed models. 