Basic Statistical Methods in Experimental Design

CET Course | SkillsFuture Claimable Course

Synopsis

MTH353 Basic Statistical Methods in Experimental Design examines how to design experiments, carry them out and analyse the data they yield. Various designs are discussed and their respective differences, advantages and disadvantages noted. Moreover, it focuses on the connection between the experiment and the model that the experimenter can develop from the results of the experiment. There are numerous examples based on real-world applications of experimental design. The course is relevant to those interested in in the design, conduct and analysis of experiments in the engineering and social sciences.

Level: 3
Credit Units: 5
Presentation Pattern: EVERY JULY

Topics

Basic Principles and Guidelines for designing experiments
Simple Comparative Experiments
Analysis of the Fixed Effects Model
Practical Interpretation of Results
Randomized Blocks, Latin Squares and Related Designs
Two-Factor Factorial Design
Fitting Response Curves and Surfaces
The General 2^k Design
Optimality of 2^k Designs
The addition of Center Points to the 2^k Design
Blocking a Replicated 2^k Factorial Design
Confounding the 2^k Factorial Design in 2^p Blocks

Learning Outcome

Determine the experimental unit, response variable, factor(s) and level(s) of a basic experiment
Demonstrate the role of randomisation and replication in inferring causation
Implement a completely randomised design
Construct the ANOVA table in R
Compute the minimum number of replicates in a completely randomised design to achieve a given level of power
Execute pairwise tests of differences in means in R to understand a significant overall F-test

Part-time Programmes / CET Courses

Singapore University of Social Sciences

Basic Statistical Methods in Experimental Design