Short Course

The validity of conclusions drawn from experimental research depends not only on a brilliant theoretical framework and meticulous experimentation, but also on reliable measurement set ups and correct application of the relevant statistics. In short – how good is the evidence that your input, your treatment, your design makes a difference? And if there is a difference –how large is it?

This short course aims to clarify the statistical background of when you can confidently state that there is a difference, and that it is linked to certain change in a parameter.

The short course will cover the following topics:

1. How good is my measurement? – Find out your Systematic error (Bias), Find out your Random error (Noise)
2. Is it the equipment or the measurement protocol? Repeatability and Reproducibility, and how to make a Standard Operating Procedure
3. When are two samples different? – 2 sample T test, 2 sample p test
4. When are 3 or more samples different? – ANOVA, Chi-square
5. What errors can we make when comparing 2 samples? Power (True Positive), Confidence (True Negative), False Negative, False Positive
6. How many repeats to take? Power and samplesize
7. How to fit a function: Regression
8. Setting up a Design Of Experiments (DOE)

Examples are shown mainly using a commercially available spreadsheet (Excel).

The short course is 9.00 – 17.30 on September 15^th. 2 coffee breaks and a course book are included in the price. Lunch is not included. A limited number of student places is available at student discount.