PrepTest 137

Incorrect. The correlation is a statement of likelihood—shoppers who checked out were more likely to buy the advertised product—so the total number of people who checked out (and the total number of people who purchased the product) are both out of scope. Picking this answer would see you fall into a percent/number trap!

Correct. Argument or Facts:
Argument

Valid or Flawed:
Flawed

Question Type:
Strengthen

Stimulus Summary:
At supermarkets, hearing an ad is correlated with buying that product. Therefore, the ads work.

Answer Anticipation:
An argument featuring a study should always make you check for both a sampling flaw and a correlation/causation flaw. Additionally, Strengthen questions feature correlation/causation flaws more frequently than any other flaw, so there's a lot pointing in that direction.

The sample here isn't really noted, so a sampling flaw is unlikely—it's much more likely when the argument says something about the sample. However, the premise uses correlative language—"more likely to"—and the conclusion uses causal language—saying something is "effective" means, well, it has an effect. So this argument features a correlation/causation ad.

Hearing an ad and buying the product in that ad are correlated, and the argument assumes that this means the ad works. However, as with any correlation/causation jump, there's always the chance of an alternative cause, a reverse relationship (unlikely here because of the timeline element), or a coincidence. To strengthen this, we should lean on our common correlation/causation strengtheners:

(1) Eliminate an alternative cause
(2) Show more examples of cause and effect going together
(3) Establish a control—shoppers who don't hear the ad and don't buy the product (note that since the statement was comparative—"significantly more likely"—there has already been a control group established so this answer is less likely)

Answer Explanation:
While this answer doesn't eliminate an alternative cause, it does eliminate an alternative possibility—the ads were just a coincidence. If most people didn't go to the store planning to buy the product, that increases the odds that the ads changed their mind. If, however, they went there to buy the products, the ads didn't have an effect and any relationship is probably coincidental.

Key Takeaway:
Probabilistic arguments—ones where something is more/less likely, or there's an increased risk, etc . . . —generally need answers stronger than some/many to have a real impact on them. A lot of Strengthen/Weaken questions fall into this pattern.

Incorrect. First, an answer about "[m]any" consumers is too weak to really impact this argument. That said, if the consumers always purchased those products, then it is more likely that any overlap between ad and purchase is coincidental.

Incorrect. First, an answer about "[m]any" consumers will have a hard time weakening a probabilistic argument. Second, since we don't have information about how memory and purchasing decisions intersect, it's impossible to know how this answer affects the argument. Maybe ads only work when people remember them; maybe they work better when they're not remembered.

Incorrect. First, an answer about "[m]any" consumers will have a hard time weakening a probabilistic argument. Second, there's no indication how frequent "occasionally" is—is it once a month, or once a year? Those would have a very different impact on the correlation. Additionally, this answer doesn't connect to the ads in any way—just the purchasing of the products.