https://blogs.timerring.com/tags/statistics/Recent content in Statistics on timerringHugoenMon, 12 Oct 2026 12:00:00 +0800https://blogs.timerring.com/posts/how-many-samples-does-it-take-to-trust-a-batch/Mon, 12 Oct 2026 12:00:00 +0800https://blogs.timerring.com/posts/how-many-samples-does-it-take-to-trust-a-batch/<p>You have 1,000 items, but testing every one of them is expensive. How many should you randomly inspect: 10, 100, or half the batch?</p> <p>There is no universal answer based on population size alone. As the <a href="https://www.itl.nist.gov/div898/handbook/prc/section2/prc22.htm" target="_blank" rel="nofollow noopener">NIST/SEMATECH e-Handbook explains</a>^{<a href="#ref-link-1" class="footnote-ref" id="ref-link-back-1">[1]</a>}, sample-size planning depends on the statistical objective and precision requirements. The useful question is: <strong>How many items do I need to estimate a proportion with a chosen level of precision?</strong> Under common assumptions, the answer is 278—not an arbitrary percentage of the batch.</p>