Implementation Models for Data Science

The choice of dataset has major implications for integration. Students can’t analyze a dataset on Water Salinity without knowing something about environmental science. If you’re a Math, Statistics, Data Science or Computer Science teacher, students are going to hit your learning goals no matter what dataset they analyze. History teachers, on the other hand, might want their students to analyze specific datasets on immigration, voting patterns, etc. A Physics teacher might want their students to collect data relating to the position of a ball rolling down a ramp over time, and a Biology teacher might want students to collect data about plants growing in the back of the classroom.

The first choice a teacher makes is what dataset(s) their students will use. Consider this first, before addressing the other implementation questions.

Depending on your subject area and grade level, you may have wildly differeny needs when it comes to data visualization, math, and statistics. A middle-school science teacher, for example, probably doesn’t need their students to confront linear regression! More than half of the Bootstrap:Data Science lessons deal with different kinds of math and statistics standards (pie and bar charts, histograms, box-plots, skew, measures of center, scatter plots, correlations, etc), but teachers should decide for themselves which lessons are important to their scope and sequence. If you’re looking to integrate Bootstrap:Data Science into your classroom, you only have to find time to teach the parts you need.

While this is the last decision you should make as a teacher, it’s also one of the most impactful. Nearly all of Bootstrap:Data Science can be taught using either a "lite" or "deluxe" programming component.

A student who completes Simple Datatypes and Contracts can do every single math and statistics lesson, exploring every type of data display and diving into measures of center and spread!

What is missing from this sequence is the ability to filter or transform their datasets, deepening their analysis and allowing for much higher engagement. But the lessons necessary to support this (Defining Functions, Table Methods, Defining Table Methods, Grouped Samples) are an extra week of class time, which not every teacher can afford.

In a CS or Data Science class, adding these lessons is a no-brainer. But for teachers integrating into Math, Science, Business, History, or Social Studies classes, this content can be left out to make the Bootstrap content take as little as a single week, or a few lessons spread out over the course of the year.