We aim to provide students with the tools, skills, and experience to work with data science. Those three components break down as such for you, the teachers of the course:
This component should be mostly handled for you, as the content of the course has been previously developed. Please familiarize yourself with the python notebooks and the high-level purposes of the libraries covered, as that is what the questions will be mostly about as you run this camp. See Course Format and Curriculum below.
Students will likely struggle if they are not familiar with the interface, the language, and the libraries. This is to be expected. The key to turning confusion into learnings in our context here is coupling their goal to accomplish the tasks with the implementation details and design. If they tell you they are stuck or can't answer a question, ask them to clarify exactly what they're trying to do, and point them to the right example or search query to address that disconnect. Providing them the answer to their question does not help nearly as much as showing them how to answer their own question
Once the students are comfortable following the motions of the content and can ask good questions about how and why things work or are designed in a certain way, you can start asking them to do more. "I see you wrote it this way, would there be another way to do that with less code?" and other kids of open-ended questions are good to ask students in the final weeks and office hour sessions. The students should have the ability to continue working on Data Science after the bootcamp, and you can help them do that by guiding them on their own problem-solving journeys and projects to explore. Providing concrete details is okay if it's done to guide the students: I know this other data set that might be fun to work with, this is another project online you might find interesting, etc.
Our curriculum is centered around two types of categories:
Tutorials are guided, step-by-step tutorials to teach concepts and technical skills. These will be video recordings with open Q&A, and they will cover the technical skills needed for the labs. These will be provided with an accompanying python notebook that should have the "Now Try This" sections completed.
Labs are self-directed assignments given between each set of tutorials to cement and test your knowledge, as well as give you some more flexibility in what you're doing with the data. For example, we will provide a dataset and perform some operations during the tutorial, but you may have another question about the data that you can answer yourself with what you've learned that week.
| Week | Name | Datasets |
|---|---|---|
| 1 | Introduction to Python & NumPy | - |
| 2 | Introduction to open data, importing data and basic data wrangling | Titanic & US Census Demographic Data |
| 3 | Introduction to data visualization and graphs with matplotlib | California Housing |
| 4 | Final project | - |
First, see course_information.md, as it discusses what the course is intended to do, the tools being covered, and the broad-strokes plan. This is assigned reading before week 1 for the students.
The week is about basic python and NumPy (Numerical Python). This covers syntax and implementation details and examples for students to look back at, as well as the ways of writing code clearly and descriptively for the moderators to provide better feedback.
NumPy is a critical data science tool for more involved data projects and scientific computing. We will not be directly using it in the following weeks, but the power and ubiquity of the package is reason enough to include it here. Using this in the final projects is a good idea!
Here we dive into the first real data science package: Pandas. Using a dataset of passenger information from the Titanic, we explore some simple correlation-style questions such as did how much somebody paid for their ticket affects their chances of survival.
This should provide a much more concrete angle to how and why data science is relevant, as well as provide the basic skills of using Pandas.
Here we get to move a bit more into displaying and building visualizations with our data. This week we introduce matplotlib to students to give control over how their data and findings are presented.
It's possible to just display an aspect or the entire data set at once, but we recognize how infrequent this is in practice. Often, you want to compare correlations between two variables or explore an individual case in more depth, and displaying that information visually can help you gain a deeper understanding as well as summing up findings to others.
We purposefully have dedicated a week of time and focus toward developing student's ideas into projects of their own. We covered three of the critical components of any good data project already: control over the code, agency to ask and answer questions about the data, and the tools to display those findings narratively.
Here is an example project we're extremely invigorated by that come to mind and that students can draw inspiration from. This is a video of the project from a data science professor at UIUC modeling recent COVID-19 case counts: https://www.youtube.com/watch?v=FSY12kiK1_o Here we see the critical components and the trial-and-error iteration that goes into a project like this, after finding a good data source. Have students brainstorm with each other, with you all as mentors leading the course, and with whomever else, they think their project could engage (professors, peers, parents, etc.).
The Bit community is powered by collaboration. We believe learning is more fun — and more effective — when we put our heads together.
We are passionate students from all walks of life who come together to making tech education accessible.
Our mission is to provide students the tools and connections they need to unleash their potential in tech.
We believe in democratizing tech education. Everyone should have access to quality resources, opportunities, and networks regardless of their background.
At Bit Project, we strive to create an environment where all people are welcomed, members are engaged, and backgrounds are celebrated.
We welcome everyone, regardless of age, race, class, ethnicity, gender identity or expression, sexual identity, ability, size, nationality, culture, faith, neurotype, and background.
We want to build a chapter community around our leaders and content. This means that at every appropriate opportunity, reaching out to younger or less experienced students is the right thing to do.
If you happened to have a good opportunity working somewhere or with somebody, pass that experience along to the group, bringing those learnings to everybody. This not only benefits others, but it builds a more supportive, trusting organization where you can learn from others the same way they learned from you.