The course provides a comprehensive basis for proper use of data in research, and is relevant to experimentalists and theoreticians alike. Topics include principles of designing an experiment, overview of data acquisition with an emphasis on avoiding general pitfalls, and how to get the information you want from the data collected. Post-acquisition (offline data analysis) procedures will cover methodological approaches, evaluation of results and their presentation. Emphasis will be on the tools and approaches required in order to obtain meaningful data, and to understand and report its significance. Examples will be used from the physical and life sciences. The course requires a rudimentary mathematical and physics background or willingness to learn the same.
The course program includes:
Mathematical background:
Fourier transforms, reciprocal space, convolution and deconvolution, error analyses (independent sources, systematic errors, etc), digitization of analog data.
Data Acquisition:
Sampling considerations, Nyquist thereom, bit noise, transfer functions, bandwidth issues, and ac (modulated) acquisition.
Data Processing (in spectroscopy and in imaging):
Noise sources and signal/noise considerations, filtering, resolution issues and object recognition, managing multi-parameter data sets.
Data evaluation:
Fitting types and their evaluation, significance, estimations, error and uncertainty analyses, introduction to deep learning.