During a quarterly quality-control audit, an engineer randomly selects 15 memory modules from a warehouse of approximately 5 000 units without replacement and records how many are defective. She plans to model the count of defectives with a Binomial(15, p) distribution to build a confidence interval for the unknown defect rate. Which fundamental assumption required by the binomial model is most likely violated by this sampling design and, if ignored, will typically over-state the sampling variance?
The total number of trials is fixed in advance at 15.
Both np and n(1 − p) must be at least 5 to justify a normal approximation.
Each trial outcome is independent of all other trials.
Every module can be classified into exactly two mutually exclusive states (defective or non-defective).
The binomial distribution assumes that every trial is independent; the outcome of one trial must not affect the probability of success on any other trial. Drawing items without replacement introduces negative dependence between draws-once a defective module is selected, the chance of selecting another defective on the next draw decreases. The correct model in this setting is the hypergeometric distribution, whose variance equals np(1 − p) multiplied by the finite-population correction (N − n)/(N − 1); this factor is less than 1, so using the binomial variance over-states the spread. The other listed conditions (binary outcome, fixed sample size) are satisfied, and the rule-of-thumb about np and n(1 − p) ≥ 5 is a guideline for normal approximation, not a core binomial assumption.
Ask Bash
Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.
Why does sampling without replacement violate the independence assumption in the binomial model?
Open an interactive chat with Bash
What is the hypergeometric distribution, and how does it differ from the binomial distribution?
Open an interactive chat with Bash
What is the finite-population correction factor, and why is it important?