Why a Random Sample Matters

Written by Studio AM.

Suppose a librarian wants to know how many books in a vast collection need repair. Inspecting every volume would take too long, so she examines a sample. If she chooses the shelf nearest her desk, however, convenience may distort the result. That shelf could contain unusually old books, popular books handled every day, or recent books that are nearly untouched.

A simple random sample uses a process that gives each book an equal chance of inclusion. Randomness does not promise a miniature copy of the library. One sample may contain more damaged books than the collection as a whole, just through chance. Its advantage is that the selection rule does not knowingly favor the convenient shelf, a certain subject, or books with visible wear. Larger random samples generally reduce chance fluctuations in the estimated proportion of damaged books.

Selection and measurement solve different problems. A perfectly random list is still unhelpful if inspectors disagree about what counts as damage. Conversely, a precise damage rule cannot rescue a sample drawn only from one unusual room.

The estimate should therefore include information about method and uncertainty. Readers need to know how books were chosen, how damage was defined, and how much sampling variation remains plausible. Readers can judge the estimate better when its selection and measurement rules are clear.

Questions

Choose an answer. The explanation appears after you answer.

  1. Question 1 of 4

    Which statement best expresses the main idea?

  2. Question 2 of 4

    Why might the shelf nearest the desk give a distorted estimate?

  3. Question 3 of 4

    What does “fluctuations” mean in the second paragraph?

  4. Question 4 of 4

    What three kinds of information should accompany the estimate?

Score: none answered yet.

More passages

Practise reading