The University of Iowa, a pioneer in Informatics research, will host a data science institute featuring five mini-classes the week of June 10, 2019. The workshop is a collaborative effort between the Iowa Social Science Research Center (ISRC) and the Iowa Informatics Initiative (UI3).
The institute will teach participants about the basic building blocks of coding as applied in the fields of the Biological, Physical, Social Sciences, and Digital Humanities. The January sessions focused on basic introductions to the software while the June sessions focus on more advanced applications using these software tools such as mixed models, social media analytics, and network analysis. We welcome faculty, staff, post-docs, graduate, and undergraduate students interested in building new skills in informatics.
Each course will involve a 3.5 to 4 hour morning session focused on instructor presentation of the material. Sessions will meet in 140 Schaeffer Hall (SH 140). The number of seats per session is limited to 70. There will be a waitlist accessed through the registration survey after the quota fills for each session.
Visit our website for more information: https://uiowa.edu/datascience.
Follow the below link to register for the June Data Science Institute. There is no registration fee, and you can register for all or individual sessions (see below).
Register here: https://uiowa.qualtrics.com/jfe/form/SV_bCJgobqHhdVilSZ
Schedule
June 10, 2019 – 8:30 Am – 12:30 Pm: Network Analysis Using R – Dr. Daniel Sewell
Introduction to Social Network Analysis with R will introduce attendees to concepts of social network analysis by illustration. The course will walk through R code, learning what the code does and introducing network concepts along the way. Attendees will leave with knowledge of commonly used R packages useful for network analysis. beginner R knowledge is recommended.
June 11, 2019 – 8:30 Am – 12:30 Pm: Mixed-effects models and related topics with R – Dr. Ariel Aloe
This course provides a practical introduction to mixed-models and related topics with R. These models allow for the analysis of nested and cross-classified data. Nested and cross-classified data structures occur often in many contexts (e.g. students nested within classrooms or schools, patients nested within clinicians, teeth nested within mouth, repeated observations nested within subject, etc). Participants will learn how to use a variety of mixed models and related packages available in R (e.g., nlme, lme4, sandwich, geepack)
June 12, 2019 – 8:30 Am – 12:30 Pm: Data visualization – static and interactive using R – Dr. Brandon Lebeau
This course will show how interactive visualizations can be created directly within R for use within web pages, R markdown documents, R notebooks, or Shiny. This hands-on workshop will provide specific examples of creating interactive visualizations using popular JavaScript libraries without needing to know JavaScript. JavaScript libraries that will be shown within R include: plotly, leaflet, DataTables, highcharts, D3. The workshop will work through examples as a group and also give time for users to create interactive visualizations on their own. Individuals will leave with example R scripts and resulting output for reference later on.
Prerequisites: Beginner R knowledge; knowledge of the tidyverse helpful, but not required. No prior JavaScript experience required.
June 13, 2019 – 8:30 Am – 12:30 Pm: Social Media Analytics with Python – Dr. Kang Lee
This unique hands-on course will cover the basics of social media analytics in Python. Participants will be able to learn how to collect, process, and analyze and visualize Twitter data using commonly-used data analytics tools and libraries in Python such as Jupyter Notebook, pandas, and networkx. Prior experience with Python is recommended, but not required.
June 14, 2019 – 8:30 Am – 12:30 Pm: Survival Analysis – Dr. Frederick Boehmke
Introduction to Survival Analysis will cover the basics of specifying, estimating, and interpreting discrete-time and continuous-time survival models. Survival models are appropriate for time-to-event data used in the social and health sciences, among other disciplines, and can be analyzed with logit and probit models, with parametric models including the Weibull, or with the Cox semi-parametric estimator. The workshop will cover many of the important features and key decisions for these models and touch on a few extensions such as competing risks, repeated failures, and cure models. Stata code and exercises will be provided and worked through – a basic knowledge of Stata will be helpful but is not essential.
