Data InsightsMulti-Source Reasoning

Free GMAT Multi-Source Reasoning Practice Question

PrepLattice is an independent test-preparation service and is not affiliated with or endorsed by GMAC, the organization that administers the GMAT. GMAT and GMAT Focus are trademarks of GMAC, used here only to name the exam this question is designed to prepare you for.

Glaze Works fires ceramic tiles in batches. Each batch gets a defect score (percentage of tiles with glaze flaws); lower is better. After one quarter the plant manager singled out 10 batches and assigned them a new slow-cool firing tweak for their next run. She reports: those 10 batches averaged a 22 percent defect rate before the tweak and only 13 percent after, so the slow-cool tweak cut defects by nine points. We should roll it out plant-wide.

Two comparison groups of batches were also tracked across the same two runs. Their mean defect rates are below. (The identity of each group and how it was formed are given in the Methodology note.)

Batch group	Mean defect rate first run	Mean defect rate second run
Comparison group 1	12 percent	12 percent
Comparison group 2	12 percent	11 percent

Glaze Works wants to estimate how much of the flagged batches' 9-point improvement is attributable to the tweak rather than to regression. For each statement, select Yes if it is a sound inference from the sources; otherwise, select No.

Statement 1: If, with no tweak, an extreme-selected group like the flagged 10 would on average regress most of the way toward the 12 percent quarter mean, then much of the 22-to-13 movement could occur without the tweak. Statement 2: Because the no-tweak random-10 group fell by only 1 point (12 to 11), the flagged group's full 9-point fall must also be only about 1 point of real effect. Statement 3: A cleaner estimate of the tweak's effect would compare the flagged batches' second-run scores against what similar extreme-selected batches scored on re-test without the tweak, not against the flagged batches' own first run.

Five fresh questions every day, your progress tracked, every miss explained. Free with an account.

Start Today's Free Set

Answer & Explanation

Correct answer

1: Yes · 2: No · 3: Yes

Statement 1: the Study summary reports the flagged 10 started at 22 percent, and the Comparison data put the quarter-wide group at 12 percent, which the Methodology note identifies as the all-batches average; because the flagged group was selected on an extreme far above that mean, regression predicts it drifts back toward the mean on re-test, so much of the 22-to-13 drop is the expected regression. Sound. Yes.

Statement 2: the Comparison data show the random group fell only 1 point, but the Methodology note says that group was not selected on an extreme, so it had almost nothing to regress; an extreme-selected group regresses much more. Transferring the 1-point movement mistakes a non-comparable baseline for a comparable one. No.

Statement 3: the right counterfactual is extreme-selected batches re-tested without the tweak, a control selected the same way the flagged group was, so both carry the same regression; comparing the flagged group only to its own first run confounds tweak with regression. Yes.