Download STATISTICAL DATA ANALYSIS

STATISTICAL DATA ANALYSIS
Course Code:
MBC7201
Course Level:
1
Course Credit:
3 CU
Brief Course Description
This course introduces students to statistical methods often used in scientific data analysis. It
begins by giving a statistical view to students and then introduces them to the basics of
descriptive statistics. The probability theory and concept, probability distributions and statistical
inference are also covered. Students are then introduced to hypothesis testing, comparison of two
mean values, basics of experimental design and one way Anova. Significance of the F test,
experiments with a block structure, factorial experiments, random and hierarchical models, splitplot experiments and checking the assumptions in Anova are also covered in this module.
Course Objectives
At the end of this course learners should be able to:
 Explain the principles underlying the various statistical methods used for data analysis
 Analyze scientific data using various statistical approaches
 Apply statistical models during their experimental design process
 Report their data collected in a scientific way backed up by statistical analysis
Course outline
Statistical view and basics of descriptive statistics
(4 hours)
Statistics and biological sciences, populations and samples and statistical inferences, making
measurements, summarizing numerical data, graphical summaries of numeric data, and
summarizing qualitative data.
Probability and probability distribution
(4 hours)
The probability concept, probability rules, discrete random variables, continuous random
variables, rules for expectations and variance and the distribution of the mean.
Statistical inference and hypothesis testing
(5 hours)
Point estimation, properties of estimators, interval estimation, testing the hypothesis, the p-value
of a test, single-sided alternative hypothesis, testing Ho: µ = µo when σ is unknown.
Comparing two mean values and basics of experimental design
(5 hours)
Inference on µ1 – µ2: matched data, inference on µ1 – µ2: principles, inference on µ1 – µ2: small
samples but equal variances, inference on µ1 – µ2: small samples but unequal variances,
inference on µ1 – µ2: large samples, Minitab example: comparing two groups; key concepts in
experimental design, examples of experimental designs
One-way ANOVA and significance of the F test
(5 hours)
Introductory example, model restrictions, subdivisions of Sums of Squares, ANOVA table,
analysis by computer; comparisons between the treatments, problems when you make many
tests, recommendations.
Block structure and factorial experiments
(3 hours)
Randomized block design, Latin Square experiments, two fixed factors in blocks (factorial
experiments), experiments with more than two fixed factors, unbalanced, and experiments.
Introductory example, model restrictions, subdivisions of Sums of Squares, ANOVA table,
analysis by computer; comparisons between the treatments, problems when you make many
tests, recommendations.
Random and hierarchical models
(3 hours)
Models with random factors (one-way Anova, two-way Anova), hierarchical models (Crossed
and nested factors, some examples).
Split-plot experiments
Introductory example, Model and Anova table, Analysis by computer.
(3 hours)
Checking the assumptions in Anova
(3 hours)
Analysis of the residues, normality, homoscedasticity, what happens if the data is not
independent, validity of the model (outliers), and residual plots in MINITAB
Tutorials
(20hours)
Mode of course delivery
This course will be conducted in three main ways i.e. formal lectures, reading
assignments/coursework, and participatory discussions/presentations.
Assessment
End of module examination, tests, assignments reports, and presentations. Their relative
contribution to the final grade is shown below:
Requirement
Assignments/presentations
Tests
Final examination
Total
Contribution
10 %
20 %
70 %
100 %
Reading List
 The recommended reading will include but not limited to the following literature.
 Ulf Olsson and Ulla Engstrand (2002). Statistics for Biologists. Swedish University of
Agricultural Sciences, Department of Biometry and Informatics
 Any other statistics text books