Repair the data-collection design · AP Statistics

Choose a sampling or assignment procedure that answers the intended question. A plan can be random in one respect and biased in another. Before choosing a procedure, state the target population, available sampling frame, response measurement and possible predictors of the response.

Build the method

A sampling frame is the list of units eligible for selection. Undercoverage occurs when the frame omits part of the target population. Nonresponse occurs after selection when sampled units do not provide usable data. Random selection from a flawed list does not repair the list.

Worked example

Label thirty people 01 through 30. Reading random pairs 01,35,01,07,00,13 selects 01,07,13 when out-of-range pairs and repeated labels are rejected. Keep reading until three distinct people are retained. Do not discard inconvenient valid labels or restart after a surprising sequence.

Compare the alternatives

For a training experiment, form novice and experienced blocks, then randomly assign both methods within each block. In contrast, assigning all novices to A and all experienced people to B makes experience and method inseparable. Blocking addresses a known predictor while random assignment supplies a fair treatment comparison inside each block.

Avoid a specific trap

Wording can bias the response even with a sound selection process. Asking about a sensible rescue plan embeds a favorable evaluation. A neutral description with support, oppose and undecided responses removes that particular pressure; it does not by itself solve undercoverage or nonresponse.

Transfer check with explanation

Transfer: if only internet users appear on the frame, adding more internet users does not give anyone offline a chance of selection. A repair must broaden the frame or narrow the stated population. If the question is causal, also specify random treatment assignment; random sampling alone is not enough.

WORKED EXAMPLE

A survey targets all local residents, but selects randomly only from a list of people with home internet. Which defect is present before anyone is contacted?

Distinguish who cannot be selected from who is selected but does not respond.
The sampling frame omits target-population members without home internet.
Why an alternative fails: Nonresponse occurs after selection when selected people do not reply; this frame excludes some before selection.

Answer: Undercoverage of people outside the frame

Compare the other answer choices

Nonresponse by sampled people

Nonresponse occurs after selection when selected people do not reply; this frame excludes some before selection.

Random assignment to treatments

There is no assigned treatment; selecting names is a sampling process.

No defect because selection is random

Random selection within an incomplete frame cannot include people absent from that frame.

Try it, then explain it.

Complete a short set without looking at the example. After each answer, explain which operation preserved the relationship and why the result makes sense.

Practice this skill

Sources & authorship

Original Slope explanations and examples. References support the concepts; they are not the source of copied questions. Independent educator review is pending.

AP Statistics Course and Exam Description, effective Fall 2026