Facebook 200-101 問題集

質問 1：
A longitudinal data set is missing values.
Which approach should be used to minimize bias in a forecast considering it is a small sample?
A. Pairwise deletion
B. Dropping variables with missing values
C. Linear interpolation
D. Winsorization
正解：C

質問 2：
A brand needs to measure how frequency affects conversions and ROI. Currently the brand runs a campaign on 2x frequency for seven days. How should the brand set up a test to understand how frequency affects incremental conversions?
A. A multi-cell A/B test: Cell 1 with a 2x frequency for seven days, Cell 2 with 4x frequency for seven days
B. A Conversion Lift with 4x frequency for seven days
C. A multi-cell Conversion Lift test: Cell 1 with a 2x frequency for seven days, Cell 2 with 4x frequency for seven days
D. A Conversion Lift test with a 2x frequency for seven days
正解：A

質問 3：
A multiple regression model establishes the relationship between Sales as a function of Ads, Price and Distribution. The analyst obtains the following output:

Which conclusion should be made about the significance of the regression?
A. Based on the R-squared, the regression equation is overall statistically significant.
B. Based on the F-test, the regression equation is overall statistically significant.
C. Based on the F-test, the regression equation is overall not statistically significant.
D. Based on the R-squared, the regression equation is overall not statistically significant.
正解：A

質問 4：
A new energy beverage brand needs to measure its upcoming digital campaign using experimental design. The campaign will run through an ad network that specializes in delivering ads to the most suitable audiences according to the message that is being communicated. When approaching a measurement partner, it learns that their suggested methodology has the following characteristics:
* Splits the intended audiences into two groups: A and B
* Group A will be shown the brand's advertising
* Group B will be shown a public service announcement instead of the brand's ad
* A survey with at least 500 respondents per group will then be conducted focusing on brand favorability
* The scores will be compared between both groups
What should the brand manager expect from this experimental setup?
A. Outcomes will be biased due to low baseline awareness
B. Outcomes will be biased because audiences in both groups are likely to be different
C. Outcome accuracy will be negatively affected due to the amount of survey responses
D. Outcome accuracy will depend on the type of public service announcement used
正解：A

質問 5：
An ecommerce brand decides to run a Facebook campaign, targeting men, to sell its recently released product. The company plans to run a single-cell Conversion Lift test to understand whether that campaign can achieve significant sales lift.
The Facebook pixel is correctly integrated on its website. It also recently released a mobile app with exclusive offers only available when a customer orders through the app. In the past, it has had a narrower target of men, ages 18-44, and it plans to use the same media weight that it has used in previous campaigns.
What is a potential issue that may affect the measurement of this campaign?
A. Using a single-cell Conversion Lift test will not accurately capture sales lift
B. Using the Facebook pixel will not track orders on the mobile app
C. Using exclusive offers will introduce bias
D. Using the same media weight on narrower targets will decrease the conversion rates
正解：A