Welcome to the ZMT Data Roundtable!
The purpose of the ZMT Data Roundtable is to provide a discussion platform about topics related to computational statistical data analysis. These topics include approaches to statistical data analysis, data handling, plotting, etc. - in R, but also in other computing languages. The Data Roundtable consists of the following two components:
- Monthly seminars: Once a month you are invited to join the Data Roundtable meetings, which can take the format of (i) a seminar, where an invited speaker presents an interesting new approach to data analysis, (ii) a trouble-shooting session, where three participants briefly (~10min) present a particular issue, followed by a discussion (~20min) to find a solution for that issue, and (iii) an open-coding session, where you are welcome to just bring your computer to work alongside your colleagues with the opportunity to ask questions whenever they arise.
- Online GitHub organization: The backbone of the Data Roundtables is a GitHub organization consisting of repositories to archive and develop the data and scripts discussed during the monthly meetings, and an issue tracker to post questions and vote, which of them should be discussed during the trouble-shooting sessions. The Github organization is private so that you may also discuss unpublished data within the Data Roundtable community. Furthermore, it provides an opportunity to ask questions and receive help even if you are not able to attend the monthly meetings.
The Data Roundtable is not a teaching tool, hosted by a few experts. It lives because of the participation of the whole data roundtable community. Anyone who is interested can contribute: The data roundtables are open to any academic level (students, doctoral candidates, postdocs, group leaders, technicians) from ZMT but also from external institutes.
First Data Roundtable in 2021: Thursday, 18.11.2021, 3:00-4:30 pm
Fridolin Haag will give a talk about "Understanding hypothesis tests as linear models".
Mini Abstract: Most of the common statistical hypothesis tests (T-test, ANOVA, ANCOVA, chi-square, etc) can be formulated as special cases of linear models (think y=a⋅x+b ). This means we can understand many tests by learning just one common framework, which may make our life as scientists much easier. In the session, I will explain the relationship between tests and linear models and we will calculate some examples to see their equivalence. You are welcome to bring cases of hypothesis testing you use in your work to discuss.
Archive of the past meetings (login required): https://github.com/ZMT-Data-Roundtable