Nonrandom Sampling Method — Definition & Example
A nonrandom sampling method is any way of choosing a sample where chance does not decide who gets picked. The researcher's judgment, convenience, or people volunteering decides instead. It is also called nonprobability sampling.
A nonprobability sampling method selects units from a population without a random mechanism. As a result, the probability that any given unit is included is unknown, or it is zero for some units. Common types include convenience sampling, voluntary response sampling, quota sampling, judgment (purposive) sampling, and snowball sampling. Because selection probabilities are not known, standard inference methods have no valid basis. These include confidence intervals and significance tests.
Worked Example
Problem: A school has 1,200 students. A student reporter wants to know whether students support a larger athletics budget. She surveys the first 50 students leaving the gym after basketball practice, and 40 of them say yes. Identify the sampling method, compute the sample proportion, and explain whether it is a trustworthy estimate for the whole school.
Step 1: Identify the method. The reporter chose students who were easy to reach at one place and time. No chance process was used, so many of the 1,200 students had zero chance of being selected. This is convenience sampling, which is a nonrandom method.
Step 2: Compute the sample proportion who said yes.
Step 3: Judge the estimate. Students leaving basketball practice are more likely than typical students to favor athletics spending. So 0.80 probably overestimates support in the school. This systematic error is called bias.
Answer: The method is convenience sampling, a type of nonrandom sampling. The sample proportion is 0.80. It is likely biased upward, so you should not use it to estimate support among all 1,200 students.
Why It Matters
AP Statistics free-response questions often ask you to name a nonrandom method and explain the direction of the bias it creates. Every inference procedure in the course lists a "random" condition, and a nonrandom sample fails it. That failure means a confidence interval or p-value from those data cannot be trusted to describe the population.
Common Mistakes
Mistake: Believing that a very large nonrandom sample, such as 10,000 online poll responses, gives an accurate result.
Correction: A larger sample reduces variability, but it does not remove bias. If the selection method favors certain people, more responses just give a more precise wrong answer. Fix the selection method by using random sampling, not by adding more people.
