Teaching Creative and Practical Data Science at Scale

Figures & data

Table 1 Required sections for submitted project reports.

Project report
1) Research question: What is the primary analysis question?
2) Hypothesis: What is the prediction?
3) Dataset(s): What data is being used to answer the question(s)? Describe the dataset(s).
4) Background: Why is this question of interest, what background information led to the hypothesis, and why is this important?
5) Methods: What methods have been used to analyze the data?
6) Ethics and privacy: Acknowledge and address any potential ethics and privacy issues related to the project.
7) Discussion: Discuss the potential civic and social impact of the project. Discuss any problems that were encountered, and enumerate the limitations of your project.

Table 2 Assignments in Data Science in Practice, and the main concepts, skills, and tools they cover.

	Concept(s)	Skills	Tools
A1	Version Control & Python	version control, python standard library	Jupyter Notebooks, git/GitHub
A2	Data Wrangling	data wrangling, tabular data	pandas, numpy
A3	Data Exploration	data wrangling, EDA, dataviz	pandas, matplotlib
A4	Data Privacy	heterogeneous data, web scraping,	pandas, requests,
		integrating datasets, reidentification	BeautifulSoup
A5	Data Analysis	data cleaning, data wrangling, EDA, regression	pandas, statsmodels, scipy, matplotlib
A6	Natural Language Processing	text data, data wrangling, TF-IDF, NLP	pandas, nltk, scikit-learn

NOTE: The main concepts, skills, and tools covered across the assignments for a single iteration of Data Science in Practice. EDA, Exploratory Data Analysis; TF-IDF, Term Frequency-Inverse Document Frequency; NLP, Natural Language Processing; nltk, Natural Language Toolkit.

Fig. 1 Screenshots from an example Jupyter notebook from a mentored group data science project from the Data Science in Practice class (Chu et al. 2019). This group analyzed publicly available crime data from the City of Los Angeles. Data were analyzed at the zip code level to examine how population, median home price, and streetlight prevalence related to crime and arrest rates.

Table 3 Ethics and privacy considerations in data science projects.

Ethics and privacy
– Do you have permission to use this data/use it for this purpose?
– Are there privacy concerns regarding your datasets that you need to deal with, and/or terms of use that you need to comply with?
– Are there potential biases in your dataset(s), in terms of who it composes, how it was collected, and when it was collected that may be problematic in terms of it allowing for equitable analysis? (For example, does your data exclude particular populations, or is it likely to reflect particular human biases in a way that could be a problem?)
– Are there any other issues related to your topic area, data, and/or analyses that are potentially problematic in terms of data privacy and equitable impact?
– How did you handle issues you identified?

Chu, J., Clinton, D., Dich, J., and Schonken, A. (2019), “COGS 108—Final Project—Crime Watch Effectiveness and Distribution,” GitHub, available at https://github.com/COGS108/FinalProjects-Fa19/blob/master/FinalProject-group030.ipynb.

 Google Scholar