Statistics with Python
This course is being designed around reasoning from data with uncertainty. No Statistics lessons, completion requirements, or assessment are published yet. Scientific Computing is a parallel pathway, not a prerequisite for this course.
Before you start
- Designed for: learners who want to interpret variation, make careful comparisons, and explain what data can and cannot support.
- You will be ready when: you have completed Python Foundations and can use functions, collections, files, and small tests to investigate a question.
- Practice setup: planned activities will run in Google Colab and will introduce the small amount of plotting or numerical-library syntax needed for each statistical investigation.
Questions the planned course will answer
Statistics will be organized around decisions a learner must justify, not a catalog of formulas. Learners will investigate questions such as:
- Which summaries describe the center, spread, and shape of this dataset?
- How can a simulation make probability and long-run variation visible?
- What changes when observations come from a sample rather than a population?
- How uncertain is an estimate, and how should that uncertainty be reported?
- Does an observed difference matter in practice as well as statistically?
- When does an association support prediction but not a causal conclusion?
- How can study design, missing data, or selection bias weaken a claim?
Planned statistical work
The course is expected to develop these connected practices:
- describe distributions with appropriate numerical and visual summaries;
- model probability through repeated experiments and simulation;
- examine sampling variability, bias, and representative data collection;
- estimate quantities with bootstrap distributions and confidence intervals;
- interpret hypothesis tests alongside effect sizes and practical importance;
- use correlation and regression as foundations for later predictive models;
- compare experiments with observational studies and identify confounding;
- communicate assumptions, uncertainty, limitations, and ethical risks.
Separate from Scientific Computing
Scientific Computing with Python will focus on numerical methods, multidimensional scientific data, specialized formats, visualization, performance, and scalable computation. Statistics will focus on how evidence from samples supports—or fails to support—a conclusion.
Neither course will require the other, so learners may complete them in either order. Both will be prerequisites for Data Science and Machine Learning with Python.
Assessment and certificate status
The planned assessment should require a reproducible analysis, correct interpretation of uncertainty, checks for misleading conclusions, and a clear explanation for a non-specialist reader. The lesson sequence, project rubric, completion threshold, and certificate requirements remain TBD. Nothing on this page currently represents a completed or verified credential.