On 30.06.2021 from 14:00-18:00 and on 02.07.2021 from 13:00-17:00 the Minds, Media Machines Integrated Graduate School (MMMIGS) is organizing an online workshop on Statistics for Human-Computer Interaction for the doctoral students of the research focus Minds, Media, Machines. The workshop will be held in English, a detailed description of the contents can be found below. Participation in both dates is recommended, but it is also possible to participate in only one of the dates.
If you are interested in participating, please send an e-mail with your name, your research area/aspired degree (e.g. computer science/doctorate) and your level of knowledge in statistics to mmmigs@uni-bremen.de.
Workshop description:
The workshop entails two sessions, each of four hours or 360 minutes. The workshop is split into two sessions to allow for reflection and deeper processing. The two sessions will be offered in close sequence to maximize retention and progress. The workshop sessions will be tailored as good as possible to the level of the participants and ample opportunities to ask questions will be provided.
The two workshop sessions can be visited separately. However, we would advise most people to visit both sessions to enable a deeper familiarization with the scientific and data driven mindset underlying the workshop.
June 30, 2020, 14:00 (online)
During the first session research methodology, data preparation and possibilities for visualization will be discussed and demonstrated. The quality of the planned research method has direct influence on the quality of the data that will be gathered according to it. Therefore, we will start with practical research methodology, enlarging upon the theoretical meaning and practical usefulness of such concepts as pre-registration, anchoring effect, double-blind setup and manipulation check. Vivid examples from real and fictitious research will be used to further the understanding of these concepts. In the following, aids to structure the data processing via R and Python will be specified and enriched with examples from own experiences. We will illustrate major challenges of a structured data preparation approach and - especially - their solutions. The end of the first session includes handy guidelines and examples for visualizing data in R (ggplot2) and Python (pandas).
July 2, 2020, 13:00 (online)
The second session will focus on analytics. Every analysis combines certain assumptions about the data. The workshop will specifically illustrate the assumptions of ANOVA, ANCOVA, (hierarchical) regression and time series analyses and their possible assumption checks. For a set of failures of certain assumptions, possibly redeeming actions are discussed and exemplified (deviation from normality, heteroscedasticity, deviation from linearity, bottom-/ceiling effects). Each analysis procedure comprises a respective set of post-hoc analyses, which are permittable and best practice, but are - unlike assumption checks - more selectively employed. As in the first session, real and fictitious examples with R and Python will enhance the active character of the workshop.
