Quick Answer: Data science is just the storytelling of numbers. You don't need Python or databases. Teach data literacy by having students collect physical classroom data, draw random samples, calculate averages, and look for misleading outliers.
Project 1: The Bias Walk
Ask students to measure the average height of students in the school. Do they sample the basketball team? The 6th grade class? Teach Selection Bias by showing how choosing a non-random sample completely ruins the data model. Then, use a Random Number Generator linked to the school roster to draw a true probabilistic sample.
Project 2: The Mean vs The Median
Write the allowance (or imaginary salary) of everyone in the class on the board. Calculate the average. Then, imagine Elon Musk walks into the room. Recalculate the average. The average is now millions of dollars. Use this to prove why the Median (the middle number) is a much better metric for skewed data.
Project 3: A/B Testing the Classroom
Introduce A/B testing (the core of tech data science). Randomly split the class in half. Give Group 1 a math worksheet printed in Times New Roman. Give Group 2 the same sheet in Comic Sans. Measure completion time. Did the font matter? You just ran a randomized controlled trial.