Udemyintermediateen

Introduction to Data Science at the Middletown Bike Rally

Explore the role of grit, genetics, and privilege in success. Understand what truly drives achievement and personal growth.

4.3(5)
520students
1h 6m
Daniel Badger Jr.
Daniel Badger Jr.Educator

Course Content

Introduction to Data Science at the Middletown Bike Rally is an intermediate-level Udemy course taught by Daniel Badger Jr.. It runs 1h 6m and is taught in English. Learners rate it 4.3 out of 5 from 5 reviews, with 520 students enrolled. It is free to enroll, last updated January 2024.

In a fascinating application of data science, 10th graders at Middletown High School analyzed data from last summer's Middletown bike rally. The project involved constructing a dataset based on the miles covered by 30 riders. Students explored the relationship between rider performance and various factors, using data analysis tools to gain insights.

The class identified three key independent variables: motivation ("grit"), aptitude ("genes"), and privilege (parental investment in equipment and training). Using questionnaire responses, each rider received scores for grit, genes, and privilege.

The students employed several data science techniques to understand how much of the variation in rider performance is attributable to these factors. They began by examining correlations between rider performance and each explanatory variable, learning to calculate and interpret correlation coefficients (positive, negative, and weak/strong correlations).

Next, the class performed simple linear regressions to assess the individual impact of each variable (grit, genes, and privilege) on rider performance. Each regression generated an equation, with a coefficient and constant, describing the relationship. They then analyzed R-squared values to determine the percentage of variance in rider performance explained by each independent variable.

To understand the combined effect of grit, genes, and privilege, the class moved on to multiple regression. This technique estimates coefficients and a constant for a single equation, incorporating all three explanatory variables to predict rider performance.

A crucial step involved understanding that future rider performance can be modeled as a random variable, a sum of random variables represented by the terms in the regression equation. They explored frequency distributions resulting from multiple trials of this random variable, observing the emergence of a normal distribution as the number of trials increased.

The class then connected frequency distributions to probability distributions. They learned to construct a normal probability distribution for a random variable using the mean (expected value) and standard error, which measures the variability around the mean.

Finally, students utilized the multiple regression equation to build a probability distribution for a rider's performance in the next rally. This allows for calculating the expected number of miles a rider will cover based on their scores, as well as the standard error of the estimate. The analysis culminated in using these probability distributions to calculate the odds of various outcomes, such as predicting the likelihood of a rider exceeding a certain mileage or surpassing a sibling's performance. This project provides hands-on experience with real-world data analysis and predictive modeling.

$ whoami — the devs behind tutorialsearch.io

Want AI in your product? Hire the team behind TutorialSearch.io

LLM features, RAG, agents and automation wired into a real product — built by the devs who shipped this site.

5.0
53m
beginner
pluralsight
5.0
1h 29m
beginner
pluralsight
5.0(766)
2h 58m
intermediate
skillshare