Offered in conjunction with Advanced Lab. I.
Introduction:
Physicists, astronomers, and generally natural scientists often work with experimental data. It is important for both experimenters and theorists to understand the nature and sources of errors, and know ways to estimate them. Experimenters must know how to calculate errors and design their methods so as to minimize them. Theorists must be familiar with the meaning of errors and with methods to compare data from various experiments as well as with theoretical model results. The Department of Physics and Astronomy offers two Advanced Laboratory courses. The first is focused on quantum and nuclear physics. The second includes research-level experiments in several areas of atomic and nuclear physics mostly involving various types of spectroscopy as well as computational modeling. All these experiments require data analysis with error calculations that typically go beyond the level of introductory physics and astronomy laboratories. Specifically, good understanding of the meaning or random errors is needed together with error propagation, chi-squared fitting, gaussian versus poissonian errors, data comparison, confidence levels. The purpose of this course is to educate students in error theory covering the needs of the Advanced Laboratory courses and to prepare them for graduate-level work in experimental physics and observational astronomy. The outcomes consist of developing the students' ability to calculate statistical errors for any experiment, propagate them, compare data with theoretical models, and draw conclusions on the accuracy and statistical significance of measured quantities.
Teacher: Athanasios Petridis
Harvey Ingham 31C
Phone: (515) 271-3723
E-Mail: Athan.Petridis@drake.edu
Class Schedule: F 3:00 pm (one standard period).
Textbook: L. Lyons, "A practical guide to data analysis for physical science students", Cambridge University Press, 1991, ISBN 0 521 42463 1 (paperback), and readings from other books
Homework: 2 assignments.
First
assignment: The students
are given a data set with reading errors and have to determine if the measured
quantities have correlated errors, calculate averages, standard deviations,
standard deviations of the mean, produce histograms, and propagate the errors
to estimate those of given functions of the data.
Second assignment: The students are given an n-tuple (a multicolumn data
table) corresponding to energy and momentum measurements of particles produced
in a high-energy experiment, produce histograms of raw and derived quantities,
fit them with gaussian distributions and determine masses of resonances
reporting on the confidence level.
Grading:
Grades are based on the two assignments.
Each assignment takes a maximum of 50 points.
100 <= points <= 85 is A
85 < points <= 75 is B
75 < points <= 65 is C
65 < points <= 50 is D
50 < points <= 0 is F
Topics:
The following topics are covered in the course
(each topic is presented and discussed in about one class period):
One class period will be reserved for discussion of each assignment.
Outcomes:
At the end of the course the students should understand statistical errors for any experiment, propagate them, compare data with theoretical models, and draw conclusions on the accuracy and statistical significance of measured quantities. This will be demonstrated by their ability to take a set of measured data and theoretical formulas for derived quantities, calculate errors on the measured data, propagate them to calculate errors of derived quantities and then compare the calculated expectation values to a theoretical model assessing the confidence level.