Learn statistics in four months
Four months of focused study — 30 minutes a day, plus a couple of weekend afternoons — teaches a numerate adult enough statistics to read empirical papers, run hypothesis tests honestly, and stop being fooled by p-values. About 70 hours total.
4 months · ~70 hours · run and interpret real tests, read studies critically
1.OpenIntro Statistics
The free textbook that has quietly replaced Mankiw-tier intro stats books at hundreds of universities. Read one chapter a week, do the end-of-chapter problems. It uses real datasets, teaches confidence intervals before NHST, and treats the reader as someone who'll actually use statistics rather than someone passing a course. The PDF and labs are free; the print copy is optional.
Free PDF; ~$25 paperback
OpenIntro Statistics →2.Khan Academy — AP Statistics
The drill layer underneath OpenIntro. Khan's exercises are short, plentiful, and force you to actually compute — most readers think they understand confidence intervals until Khan asks them to construct one. Use it the day after each OpenIntro chapter. The unit tests are the honest checkpoint; if you fail them, reread.
Free
Khan AP Statistics →3.StatQuest — Josh Starmer
When a concept refuses to click — bootstrapping, ANOVA, logistic regression — search "StatQuest [topic]." Josh Starmer explains every standard statistical method with cartoons, jingles, and a refusal to hide the math behind hand-waving. His videos are the single best supplemental resource in the field. Don't binge them; use them as a tool.
Free
StatQuest YouTube →If this doesn't fit you
If you only need statistics to interpret studies — not to run them — skip the textbook and read Wheelan's Naked Statistics in a weekend, then watch StatQuest's intro playlist. Two weeks instead of four months. You will not be able to compute, but you will not be fooled by a misused chi-square test either.
Why this path
Most intro stats courses fail by teaching procedures without intuition, leaving graduates who can run a t-test and have no idea what they just did. OpenIntro inverts the order — concept, then formula. Khan keeps the muscle memory honest. StatQuest catches you when both fail. Skipping the problem sets is the most common mistake; statistics rewards reps the way calculus does, and watching is not learning.