Course Identification

This course aims to familiarize students with executing a variety of statistical tests: t-tests, ANOVA and its different variations, multiple linear regression and more. The principle aim of the course is to enable students to understand which analysis is applicable for each type of data, and execute the proper analyses using R. The main focus will be on usability and application of statistical knowledge in answering research questions, and less on the mathematical background of the statistical methods. No background in programming is needed.

Topics by week:

    Introduction to R
    Descriptive statistics
    Comparing two populations: t-test.
    Comparing two populations - non-parametric tests
    One-way ANOVA
    Multiple comparisons and contrasts
    Two-way ANOVA
    Experimental design - randomized block, nested, split-plot
    Experimental design - repeated measures
    Data transformation, power calculations
    Linear regression
    Correlation + selected topics

Upon successful completion of this course students should be able to:

    Determine the appropriate statistical analysis for a given experimental setting.
    Plan experiments while considering statistical power, sufficient sample size, and avoiding potential statistical "pitfalls".
    Execute a variety of common statistical tests (e.g., t-tests, ANOVA and its different variations, multiple linear regression, non-parametric analysis) using R (although the knowledge obtained will be applicable for commercial statistical software as well).
    Critically review statistical analyses conducted in published papers.

Sokal, Robert R. Biometry: the principles and practice of statistics in biological research.
Zar, Jerrold H. Biostatistical Analysis
Jolicoeur, P. Introduction to Biometry (available as e-book).
Software: The course will use R, which is freely accessible. Commercial statistics packages you can use on your own include Statistica, Systat, SPSS, JMP, and Stata. Less user-friendly options are SAS, which is available at Weizmann via a site license, and Matlab.