A parcel carrier studies the rule that a fragile mark implies extra padding. Among shipments already marked fragile, the planner wants how often extra padding is present too. Which association measure is that?
Select an answer to reveal the explanation.
Short Explanation
Confidence is the follow-up question: among the fragile boxes, how often did the padding show up too. That conditional rate is what you want, not the share of the whole warehouse. The trap is support, which counts the pair across every shipment and does not condition on the fragile mark.
Full Explanation
Confidence of a rule is how often the consequent appears when the antecedent is already present. Here the antecedent is the fragile mark and the consequent is extra padding, so confidence is the padding rate computed only among shipments already marked fragile. That is the planner's question. Support is the joint share of fragile and padding across all shipments, with no conditioning on the fragile mark, so a large pile of ordinary parcels pulls it down even when fragile parcels are usually padded. Lift divides confidence by how often padding appears overall, which answers whether the rule beats the base rate rather than stating the conditional rate itself. Antecedent support counts fragile shipments whether or not padding was added, so it ignores the consequent completely. The denominator is the set of fragile shipments, not every parcel the carrier handled. Exam caveat: a high confidence on a rare antecedent can still describe very few shipments, so confidence does not replace a look at how common the pair is. Operational check: filter to fragile shipments only, divide the padded ones by that filtered count, and confirm the result is the reported confidence.