explanation.md

* Tom has a true weight of 70. He stands repeatedly on his scale to measure his weight. * The scale might be erroneous and return biased measurement values.
* If the **measurement error is (only) random**, each value is biased, however, the mean across repeated measurements equals Tom's true weight. The average bias across measurements is 0. * If the **measurement error is systematic**, each value gets a certain bias. As a consequence, the mean across repeated measurements is biased and does not equal the true value. * If a **combination of both random error and systematic error** occurs, all measurements are systematically shifted away from the true value besides "individual" random biases in different directions.
* Questions for discussion in class + What is the random error and what is the systematic error. Discuss! + How can we extend these ideas when we collect data on a sample of different individuals rather than having repeated measurements on the same individual? + What would this graph look like if the systematic error you choose would suddenly change after 50 measurements? How could that happen for the scale example?
* **Resources** + [This article](https://www.nde-ed.org/GeneralResources/ErrorAnalysis/UncertaintyTerms.htm) provides a nice explanation regaring various concepts. + There are wikipedia pages on [random error](http://en.wikipedia.org/wiki/Random_error) which is closely linked to the idea of the [precision of a measure](http://en.wikipedia.org/wiki/Accuracy_and_precision) and on [systematic error](http://en.wikipedia.org/wiki/Systematic_error) which is linked to the concept of [accuracy](http://en.wikipedia.org/wiki/Accuracy_and_precision).
* Issues: Measurements in the Bull's eye plot should lie on a single point (not small circle..).