Introduction to Data Science, case-by-case

Interest in data science is surging, which means nowadays it’s pretty easy to fill the seats in an introductory data science class. But how do we effectively take the students through a challenging curriculum once they are in the class? We argue that the answer is an application first approach where the curriculum is divided into learning modules, each covering a batch of connected learning goals and designed around a case study. In this talk we present the curriculum for such a course intended for an audience of Duke University students with little to no computing or statistical background, and focuses on data wrangling, exploratory data analysis, data visualization, and effective communication. This course serves not as a first and thorough exposure to computing essentials for data science (including programming with R, reproducibility with R Markdown, and version control and collaboration with git/GitHub) but also as a gateway for the statistical science major. We will discuss in detail the course design philosophy and pedagogical considerations as well as give examples from the case studies used in the course.

Transcript

1. mine-cetinkaya-rundel cetinkaya.mine@gmail.com @minebocek mine çetinkaya-rundel duke university + rstudio intro to data science: case by case bit.ly/enar19-cases
2. FIRST YEAR Introductory courses: survey of methods / tools / approaches SENIOR Capstone / case studies SOPHOMORE Intermediary courses: deeper look at fundamentals JUNIOR Applied and theoretical electives bit.ly/enar19-cases typical undergraduate curriculum
3. QHow can we design an introductory curriculum around case studies? bit.ly/enar19-cases
4. APPLICATION FIRST BATCH OF CONNECTED LEARNING GOALS DESIGNED AROUND CASE STUDIES learning modules bit.ly/enar19-cases
5. QWhy should we consider designing an introductory curriculum around case studies? bit.ly/enar19-cases
6. illustrate process provide motivation dem onstrate am biguity
7. Q Which of the following is more likely to be motivating for a wide range of students? bit.ly/enar19-cases
8. option 1: ✓ Topic: Web scraping & mapping ✓ What will we learn about? ✓ rvest: A new package for harvesting data off the web ✓ regular expressions ✓ ggplot2’s mapping features ✓ Functions and automation bit.ly/enar19-cases
9. option 2: ✓ Case study: Money in Politics regex fn bit.ly/enar19-cases
10. Money in politics OPENSECRETS.ORG bit.ly/enar19-cases
11. students will encounter lots of new challenges along the way — let that happen, and then provide a solution bit.ly/enar19-cases
12. Lesson: Web scraping essentials for turning a structured table into a data frame in R. bit.ly/enar19-cases
13. Lesson: Web scraping essentials for turning a structured table into a data frame in R. Ex 1: Scrape the table off the web and save as a data frame. bit.ly/enar19-cases
14. Lesson: Web scraping essentials for turning a structured table into a data frame in R. Ex 1: Scrape the table off the web and save as a data frame. Ex 2: What other information do we need represented as variables in the data to obtain the desired facets? bit.ly/enar19-cases
15. Lesson: Web scraping essentials for turning a structured table into a data frame in R. Ex 1: Scrape the table off the web and save as a data frame. Ex 2: What other information do we need represented as variables in the data to obtain the desired facets? Lesson: “Just enough” string parsing and regular expressions to go from to bit.ly/enar19-cases
16. illustrate process provide motivation dem onstrate am biguity
17. bit.ly/enar19-cases
18. FIND A DATASET PIN DOWN A RESEARCH QUESTION IDENTIFY METHODS AND TECHNIQUES PERFORM ANALYSIS WRITE REPORT / PREPARE PRESENTATION how we might expect students to approach a final project… bit.ly/enar19-cases
19. FIND A DATASET PIN DOWN A RESEARCH QUESTION IDENTIFY METHODS AND TECHNIQUES PERFORM ANALYSIS WRITE REPORT / PREPARE PRESENTATION …how students might approach a final project bit.ly/enar19-cases
20. Poverty index HOUSEHOLD ASSET HOLDINGS + DEMOGRAPHICS FROM VARIOUS COUNTRIES bit.ly/enar19-cases
21. there is no one clear answer, allow students to brainstorm approaches and take them through your (expert) reasoning for what might / might not work bit.ly/enar19-cases
22. illustrate process provide motivation dem onstrate am biguity
23. ASA DataFest is a data analysis competition where teams of up to five students attack a large, complex, and surprise dataset over a weekend. inspiration… bit.ly/enar19-cases
24. Paris Paintings PAINTING AUCTION DATA 1764 - 1780 bit.ly/enar19-cases
25. introduce students to the formulation of research questions, and help them understand what questions can (and cannot) be answered with a given dataset bit.ly/enar19-cases
26. Two paintings very rich in composition, of a beautiful execution, and whose merit is very remarkable, each 17 inches 3 lines high, 23 inches wide; the first, painted on wood, comes from the Cabinet of Madame la Comtesse de Verrue; it represents a departure for the hunt: it shows in the front a child on a white horse, a man who gives the horn to gather the dogs, a falconer and other figures nicely distributed across the width of the painting; two horses drinking from a fountain; on the right in the corner a lovely country house topped by a terrace, on which people are at the table, others who play instruments; trees and fabriques pleasantly enrich the background. bit.ly/enar19-cases
27. bit.ly/enar19-cases
28. QWhat are resources for case studies for an introductory curriculum? bit.ly/enar19-cases
29. data expeditions PAIR OF GRAD STUDENTS, WORK WITH COURSE INSTRUCTOR TO FORMULATE A QUESTION, + A PATHWAY THROUGH A DATASET TO EXPLORE THE QUESTION ELEMENT OF AN UNDERGRADUATE COURSE THAT INTRODUCES STUDENTS TO EXPLORATORY DATA ANALYSIS GRADUATE STUDENT PARTICIPANTS RECEIVE A TRAVEL GRANT bit.ly/enar19-cases
30. datasciencebox.org bit.ly/enar19-cases
31. Visualizing data Wrangling data Making rigorous conclusions Looking forward Fundamentals of data & data viz, confounding variables, Simpson’s paradox + R / RStudio, R Markdown, simple git Tidy data, data frames vs. summary tables, recoding and transforming, web scraping and iteration + collaboration on GitHub Building & selecting models, visualizing interactions, prediction & validation, inference via simulation Data science ethics, interactive viz & reporting, text analysis, Bayesian inference + communication, dissemination bit.ly/enar19-cases
32. mine-cetinkaya-rundel cetinkaya.mine@gmail.com @minebocek mine çetinkaya-rundel duke university + rstudio intro to data science: case by case bit.ly/enar19-cases