Skip to content

ACCA Strategic Professional · Advanced Performance Management · Data science and analytics

Marlow Group links sales managers' bonuses to a customer satisfaction score from automated surveys. Analytics shows that scores rose 15% after managers began sending surveys only to customers who had just received discounts, while repeat purchases fell. Which conclusion best reflects data-driven reporting and ethical risks?

Selection bias and gaming are distorting the satisfaction score, as shown by rising scores but falling repeat purchases. The reward scheme should be redesigned with controlled sampling and balanced with outcome measures. Automatic collection does not make a biased sample reliable.

  1. AThe measure is reliable because the score rose and is automatically collected
  2. BSelection bias and gaming distort the metric; the reward link should be redesigned with controls over sampling and balanced with outcome measures such as repeat purchasesCorrect
  3. CSampling should be reduced further to lower survey costs
  4. DThe fall in repeat purchases proves the survey software is inaccurate

Explanation

Surveying only discounted customers is selection bias, and managers are gaming a metric tied to pay, so the score no longer reflects true satisfaction. Falling repeat purchases confirms the divergence. The fix is controlled random sampling and balancing with outcome measures; automation of collection does not guarantee representativeness.

Did you get it right without looking?

One question tells you little. A timed set on Data science and analytics shows your real accuracy, how long you take and where you lose marks.

Practise Data science and analyticsFree diagnostic mock

More Data science and analytics questions