Biostatistics Course

Lesson 1 • Role of Statistics in Health Research

Defines biostatistics and its function in evidence-based medicine. Establishes why quantitative reasoning is essential to clinical and public health decisions.

Lesson 2 • Organising and Displaying Data

Covers frequency tables, histograms, and bar charts for initial data exploration. Visualisation reveals distributional patterns before formal analysis.

Lesson 3 • Data Collection and Study Design Basics

Introduces observational and experimental data sources and their limitations. Links collection method to potential bias and data quality.

Lesson 1 • Exploratory Data Analysis in Practice

Applies descriptive tools to real health datasets to detect anomalies and patterns. EDA precedes formal inference and guides hypothesis generation.

Lesson 2 • Measures of Central Tendency

Teaches mean, median, and mode with emphasis on when each is appropriate. Connects choice of measure to variable type and distributional shape.

Lesson 3 • Summarising Data with Box Plots

Constructs and interprets box plots to display five-number summaries. Box plots efficiently communicate dispersion, median, and outliers simultaneously.

Lesson 4 • Measures of Variability

Covers range, variance, standard deviation, and interquartile range. Variability measures quantify uncertainty and data dispersion around the centre.

Lesson 1 • Conditional Probability and Bayes Theorem

Explains conditional probability and updates beliefs using Bayes theorem. Directly applicable to diagnostic test interpretation and disease screening.

Lesson 2 • Basic Probability Concepts

Defines probability, sample spaces, and events using health examples. These concepts are the logical foundation for all statistical inference.

Lesson 3 • Discrete Probability Distributions

Covers binomial and Poisson distributions with health-relevant parameters. Models count outcomes such as disease cases and adverse events.

Lesson 4 • Continuous Probability Distributions

Introduces the normal and t-distributions and their probability density functions. Continuous distributions model biological measurements and sampling variability.

Lesson 1 • Point Estimation Principles

Defines estimators and desirable properties such as unbiasedness and efficiency. Establishes the conceptual link between sample statistics and population parameters.

Lesson 2 • Confidence Intervals for Means

Constructs z-based and t-based intervals for single and paired means. Interval width reflects sample size and variability in the data.

Lesson 3 • Sample Size Determination for Estimation

Derives formulas to achieve desired margin of error for means and proportions. Proper sample size planning ensures studies are adequately powered and efficient.

Lesson 1 • Logic and Framework of Hypothesis Testing

Explains null and alternative hypotheses, significance levels, and decision rules. The framework structures how evidence is evaluated against a default assumption.

Lesson 2 • Tests for Categorical Data

Applies chi-square tests for independence and goodness-of-fit to frequency tables. These tests evaluate associations between categorical variables in health studies.

Lesson 3 • Statistical Power and Sample Size for Tests

Defines power, beta error, and effect size and links them to sample size. Adequate power prevents false-negative conclusions in clinical research.

Lesson 4 • Two-Sample Tests for Means

Compares means from two independent groups and paired observations. Selecting the correct test depends on study design and variance assumptions.

Lesson 1 • Correlation Measures and Interpretation

Calculates Pearson and Spearman correlation coefficients and tests their significance. Correlation quantifies linear and monotonic associations without implying causation.

Lesson 2 • Regression Diagnostics and Assumptions

Checks linearity, homoscedasticity, normality of residuals, and influential points. Violated assumptions bias estimates and invalidate inference.

Lesson 3 • Simple Linear Regression

Fits a straight-line model to predict a continuous outcome from one predictor. Regression coefficients quantify the magnitude and direction of the association.

Lesson 4 • Multiple Linear Regression

Extends regression to multiple predictors and introduces confounding control. Adjusted coefficients isolate each predictor's independent contribution to the outcome.

Lesson 1 • One-Way Analysis of Variance

Partitions total variability into between-group and within-group components. ANOVA tests whether at least one group mean differs from the others.

Lesson 2 • Repeated Measures and Mixed Designs

Handles within-subject measurements over time or conditions using repeated measures ANOVA. Accounts for correlation among repeated observations to improve power.

Lesson 3 • Two-Way ANOVA and Interaction Effects

Analyses two categorical factors simultaneously and tests for interaction. Interaction reveals whether one factor's effect depends on the level of another.

Lesson 4 • Nonparametric Tests for Ordinal and Non-Normal Data

Applies rank-based alternatives to t-tests and ANOVA when normality is untenable. Nonparametric tests preserve validity with small samples or skewed distributions.

Lesson 1 • Cox Proportional Hazards Regression

Models the effect of covariates on the hazard rate using the Cox model. Hazard ratios from this model adjust for confounders in survival studies.

Lesson 2 • Introduction to Survival Analysis

Defines censoring, survival functions, and hazard functions for time-to-event data. Survival methods handle incomplete follow-up that standard regression cannot address.

Lesson 3 • Standardisation and Adjustment of Rates

Removes confounding by age or other factors using direct and indirect standardisation. Adjusted rates enable valid comparisons across populations with different compositions.

Lesson 4 • Measures of Disease Frequency

Defines incidence, prevalence, and mortality rates and their denominators. Accurate frequency measures are the basis for comparing disease burden across populations.