Applied data analysis for psychology using the open-source software R [Data Analysis using R]

Dozent/in:

Alexander Pastukhov

Angaben:

Seminar, 2 SWS, ECTS: 3

Termine:

Do, 12:00 - 14:00, M3N/-1.19

Voraussetzungen / Organisatorisches:

No programming background necessary, basic knowledge on statistical analysis is advantages but not strictly necessary.

Inhalt:

An introductory hands-on course that shows how to use R to analyze a typical psychophysical and social psychology research data. The course will walk you through all the analysis stages from importing a raw data to compiling a nice looking final report that automatically incorporates all the figures and statistics. If description below looks intimidating, do not despair! R wraps all these steps into simple easy-to-understand procedures. These include:

• data import, whether it is a CSV-file, Excel, SPSS, SAS, or Matlab and merging multiple data files into a single easy-to-use table (and saving it)
• data preprocessing: filtering out bad data and transforming values, e.g. turning continuous data into ordinal, degrees to radians, skewed data to a normally distributed ones, renaming and relabeling conditions, working with dates, etc.
• grouping and summarizing data: grouping data based on various combinations of conditions, computing group statistics or transforming data within each group
• plotting: R plotting package ggplot2 makes exploratory analysis easy and generates production-quality figures that you can use directly for your thesis or a publication
• statistical analysis: you will learn how to use various parametric (ANOVA, t-test, linear-mixed models) and non-parametric (traditional and permutation based) statistical tests, as well as the Bayesian approach to statistics and modern predictive modelling approaches (regression and classification using generalized linear models, linear discriminant analysis, support vector machines, etc.).
• putting it all together into a report: R helps you to automatically generate a nice looking report that includes all the analysis steps, figures, and statistics and export it for direct presentation (PDF, HTML) or for further editing (Word). Best part, if you change your mind and analysis, your final looking report is one button press away.

The best way to attend the course is with your own dataset. Bring it to the course and see how R will allow you to understand it deeper or reduce the time you need to analyze it. I am happy to help you all along with such statistical, methodological and graphical problems.

Empfohlene Literatur:

"R for Data Science" by Garrett Grolemund and Hadley Wickham available freely at http://r4ds.had.co.nz/

Schlagwörter:

statistical analysis, data science, statistics

Empiriepraktikum Consciousness and Multi-stable Perception [EMPRA Consiousness]

Dozent/in:

Alexander Pastukhov

Angaben:

Seminar, 4 SWS

Termine:

Di, 14:00 - 18:00, MG2/01.09

Inhalt:

We are used to the fact that our visual perception is stable and unambiguous. However, so-called multi-stable displays such as Necker cube or Binocular Rivalry are consistent with several comparably likely interpretations. This forces our visual system to endlessly switch between these alternatives despite the constant display, making them one of the favorite tools to study consciousness. They are also are highly revealing as they show the inner workings of sensory representations, attention, priming, prior knowledge, short-term memory, etc. The aim of the practicum is to:

• Introduce the current research on multi-stable displays and on their use in studying both perception and consciousness.
• Decide on your own research topic. I have several paper-worthy ideas to start from (visual-audio interactions, perceptual coupling between several copies of displays, perception of invisible Necker cubes, EEG, …) but you will have an opportunity develop them to your liking.
• Design an experiment. I will guide you through this process, showing you best strategies for measuring subjective conscious perception and maximizing your chances for interpretable data.
• Collect data, analyze it and turn it into beautiful evidence. I will explain how to turn the raw data into a simple clear and appealing evidence (taking part in Data Analysis using R seminar is highly recommended but not required). And, it must look beautiful, so we will explore the rules of beautiful evidence proposed by Edward Tufte.
• Present you work at the Empiriepraktikumskonferenz!