Data Science Foundations

What you will be able to do

• You will be able to scope a vague business question into a data science problem with a defined target, unit of analysis, and success metric, using a lifecycle like CRISP-DM.
• You will be able to summarize a dataset honestly with the right descriptive statistics (central tendency, spread, and shape) and identify the probability distribution it follows.
• You will be able to reason about uncertainty with sampling, the central limit theorem, and confidence intervals rather than treating a single sample as the whole truth.
• You will be able to run a hypothesis test, read the p-value and effect size correctly, and state plainly what the result does and does not prove.
• You will be able to compute efficiently on numeric data with numpy arrays, broadcasting, and vectorized operations instead of slow Python loops.
• You will be able to load, clean, reshape, join, and aggregate messy real-world data into an analysis-ready pandas DataFrame.
• You will be able to run a structured exploratory data analysis that surfaces outliers, missing data, and relationships across variables using summary tables and plots (histograms, box plots, and scatter plots).
• You will be able to read correlation without mistaking it for causation, and communicate your findings in plain language and clear visuals that support a decision.

What is inside

4 modules, 8 lessons. Each module ends in a graded quiz and most carry an assignment.

1. 01

   The Data Science Process

   Most data science projects are lost at the framing stage, not the modeling stage. This module walks the full lifecycle from a business question to a measurable outcome using CRISP-DM, then establishes the data-type literacy and reproducible, notebook-based habits that keep your results trustworthy.

   2 lessons · 5 quiz questions

2. 02

   Statistics and Probability You Need

   Every chart, test, and model rests on a small set of statistical ideas, and getting them wrong quietly corrupts everything downstream. This module builds the descriptive statistics, probability, and distributions you use daily, then moves into sampling and inference so you can separate real signal from noise, summarize data honestly, and state exactly what a test result proves.

   2 lessons · 5 quiz questions · assignment

3. 03

   Python for Data: numpy and pandas

   numpy and pandas are the daily workbench of a working data scientist, and real fluency in them is what separates fast analysts from stuck ones. This module opens with numpy arrays, broadcasting, and vectorized math, then walks through loading, cleaning, reshaping, joining, and aggregating real datasets with pandas. By the end, messy source files turn into tidy, analysis-ready tables you can actually explore.

   2 lessons · 5 quiz questions

4. 04

   Exploratory Data Analysis

   Exploratory data analysis is where a clean table turns into questions, hypotheses, and decisions, and it is the skill stakeholders actually see you use. This module builds the core moves: univariate exploration with histograms and box plots, bivariate exploration with scatter plots, and reading correlation without being fooled by nonlinearity, outliers, or confounders. You will also learn to handle the two problems that quietly wreck analyses, outliers and missing data, and to classify missingness so your fixes are defensible. You finish by turning what you found into plain-language findings and clean visuals a decision-maker can actually act on.

   2 lessons · 5 quiz questions · assignment