Statistical Methods in Experimental Science2010VT
|
|
Course plan
Learning objectives
Having completed the course, the student shall be able to:
- display good understanding of major principles for statistical analysis
of experimental data;
- explore a given set of experimental data in search of a suitable
statistical model;
- interpret the results of major types of statistical analysis of
experimental data.
Contents
Descriptive statistics: mean, median, standard deviation, boxplots.
Probability theory: probability distributions, binomial, Poisson and normal
distributions.
Statistical inference: Point estimators, confidence intervals, hypothesis
testing.
One-, two- and multisample tests for means and medians: t-tests, one-way and
two-way analysis of variance; block designs; rank sum tests, Kruskal-Wallis and
Friedman tests.
Regression and correlation: Pearson and Spearman correlation; Simple and
multiple linear regression. Polynomial regression. Dummy variables. Model
selection. Covariance analysis. Models of repeated measures.
Admission requirements
The student must have passed a minimum of 4.5 ECTS credits in statistics and 15
ECTS credits in statistics and mathematics together. This corresponds to what
is normally included in biology and chemistry programs at LiU.
The course will be given in English. The course includes lessons, 2-3 hours
each, at 6-10 occasions.
Teaching
The course is problem-oriented, and the student is expected to bring relevant datasets from his/her research field.
Examination
Written and oral presentations of solutions to assignments encompassing statistical analysis of experimental data.
Course literature
Hassard, T.K. Understanding biostatistics. ISBN 0-8016-2078-3.
Examiner
Anders Grimvall
Phone: 013-28 1482, e-mail: angri@mai.liu.se
Teaching period
April 26 to June 4 2010 (exact start and end date will be decided by the participators and the examiner).
Credits
3-4.5 ECTS credits
Organized by
Forum Scientum
Page responsible: Director of Graduate Studies