References: How sure are we? confidence intervals

Source material

Source curriculum (structural mirror, cited as further study):
• Khan Academy, "Confidence intervals" (Statistics & Probability)
  Author: Sal Khan and the Khan Academy team
  Unit page: https://www.khanacademy.org/math/statistics-probability/confidence-intervals-one-sample
  License: CC BY-NC-SA 4.0
Clawdemy's lessons are original prose that follows the pedagogical arc of this
unit. We do not embed, reproduce, or transcribe Khan's text or videos; we link
out to the relevant unit as recommended further study. The non-commercial
clause aligns with Clawdemy's free, zero-revenue posture. All rights to the
original materials remain with their authors.

Source-scope note: this lesson mirrors Khan's "Confidence intervals" unit
(margin of error, the meaning of a confidence level, the correct
interpretation) and restates it in Clawdemy's voice with original examples
(model accuracy, conversion rate). The standard error it relies on comes from
the previous lesson. The AI framing (reporting metrics with intervals,
overlapping intervals signalling indistinguishable results) is Clawdemy
framing. Exact per-unit URLs are verified at promotion.

Read this next

Khan Academy: Confidence intervals by Sal Khan and the Khan Academy team. The full unit this lesson mirrors, with videos and practice on margins of error and the correct interpretation, free and CC-licensed. Especially worth it for drilling the interpretation, which is the part that trips people up.

Going deeper

A short, durable list. Both are free.

Khan Academy, “Sampling distributions” (within the course above). The previous lesson’s source; revisit the standard error, which is the engine inside the margin of error here.
Khan Academy, “Significance tests (hypothesis testing)” (within the course above). The direct next step: instead of a range around one estimate, deciding whether an observed difference is real. This is Track 9’s next lesson, and confidence intervals and tests are two views of the same idea.

Adjacent topics

Where this sits inside this track.

From sample to population: sampling and the central limit theorem. The previous lesson. The standard error and the normal shape it provides are exactly what make the interval work.
Testing a claim: hypothesis testing and p-values. The next lesson. Overlapping confidence intervals hint that a difference may be noise; the hypothesis test makes that call formally.
Statistics in machine learning. The capstone. Confidence intervals on metrics are part of evaluating models honestly, which the capstone ties together.