A marketing firm claims that a new advertising campaign increased product sales by 20%. Critics argue that the increase in sales might not be solely due to the campaign, pointing out that a major competitor went out of business shortly before the campaign launched. Which of the following strategies would best help the marketing firm demonstrate that the advertising campaign, rather than the competitor's exit, was the primary cause of the sales increase?
- AAnalyze sales data from regions where the advertising campaign ran but the competitor did not operate.
- BCompare the sales increase to the company's average sales growth over the past five years.
- CConduct a customer survey to determine brand awareness and recall of the new advertising campaign.
- DExamine the competitor's market share and product overlap with the firm's product before its exit.
Show answer & explanationAnswer & explanation
Correct answer: A. Analyze sales data from regions where the advertising campaign ran but the competitor did not operate.
The critics suggest a confounding variable (competitor's exit). To isolate the campaign's effect, the firm needs a control group where the campaign was present but the confounding variable was not. Analyzing sales in regions with the campaign *but no competitor* would help control for the competitor's exit, allowing for a more accurate assessment of the campaign's impact.
Why the other options are wrong
- B. Comparing to historical average growth doesn't account for the unique external factor of the competitor's exit.
- C. Customer surveys measure campaign effectiveness but don't isolate it from the competitor's exit as a cause for sales increase.
- D. This helps understand the competitor's impact but doesn't isolate the campaign's effect from that impact.
Controlling for Confounding Variables
Controlling for confounding variables involves designing a study or analysis to minimize the influence of extraneous factors that could obscure the true relationship between an independent and dependent variable.
- Confounding variables affect both cause and effect.
- Methods include randomization, matching, statistical control.
- Helps establish a clearer causal link.
Memory trick: Infer the cause, clear the noise and flaws.