Scribd
Upload
7 views5 pages

Ab Testing Problems

The document describes 4 problems related to A/B testing. For each problem, it provides data for two groups (A and B) testing different variables. It then calculates metrics to compare the groups and determine which variable is more effective based on the data. For problem 1, it compares like rates between a new social media algorithm and chronological order. For problem 2, it compares enrollment rates between new and current education interfaces. For problem 3, it compares average order amounts between groups with and without a discount coupon. For problem 4, it compares average calories burned between two different workout routines.

Uploaded by

Syed Asad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
7 views5 pages

Ab Testing Problems

The document describes 4 problems related to A/B testing. For each problem, it provides data for two groups (A and B) testing different variables. It then calculates metrics to compare the groups and determine which variable is more effective based on the data. For problem 1, it compares like rates between a new social media algorithm and chronological order. For problem 2, it compares enrollment rates between new and current education interfaces. For problem 3, it compares average order amounts between groups with and without a discount coupon. For problem 4, it compares average calories burned between two different workout routines.

Uploaded by

Syed Asad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

A/B Testing Problems

Problem 1:

A social media platform wants to test a new algorithm for displaying posts to users in their news
feed. They randomly select two groups of users: Group A sees posts in chronological order, while
Group B sees posts using the new algorithm. After a week of testing, the platform collects the
following data:

Group A:

- Number of impressions: 10,000

- Number of likes: 800

Group B:

- Number of impressions: 9,500

- Number of likes: 950

Calculate the like rate (percentage of impressions that resulted in likes) for both groups and
determine which approach is more effective.

Answer:

Like Rate for Group A:

Like Rate = (Number of Likes / Number of Impressions) * 100

= (800 / 10,000) * 100

= 8%

Like Rate for Group B:

Like Rate = (Number of Likes / Number of Impressions) * 100

= (950 / 9,500) * 100

= 10%
Conclusion:

The like rate for the new algorithm (Group B) is 10%, which is higher than the like rate for the
chronological order (Group A) at 8%. This suggests that the new algorithm is more effective in
generating likes on posts.

Problem 2:

An online education platform is considering changes to its user interface to increase user
engagement. They randomly select two groups of users: Group A experiences the current interface,
and Group B experiences the updated interface. After two weeks of testing, the platform collects
the following data:

Group A:

- Number of course enrollments: 500

- Number of active users: 1,000

Group B:

- Number of course enrollments: 550

- Number of active users: 900

Calculate the enrollment rate (percentage of active users who enrolled in a course) for both groups
and determine if the updated interface has a significant impact.

Answer:

Enrollment Rate for Group A:

Enrollment Rate = (Number of Course Enrollments / Number of Active Users) * 100

= (500 / 1,000) * 100

= 50%

Enrollment Rate for Group B:


Enrollment Rate = (Number of Course Enrollments / Number of Active Users) * 100

= (550 / 900) * 100

= 61.11%

Conclusion:

The enrollment rate for the updated interface (Group B) is 61.11%, which is higher than the
enrollment rate for the current interface (Group A) at 50%. This suggests that the updated interface
has a positive impact on user engagement and course enrollments.

---

Problem 3:

A food delivery app wants to test the effectiveness of a new discount coupon on increasing order
sizes. They randomly select two groups of customers: Group A receives no coupon, and Group B
receives a 20% discount coupon. After a week of testing, the app collects the following data:

Group A:

- Total order amount: $10,000

- Number of orders: 500

Group B:

- Total order amount: $12,500

- Number of orders: 600

Calculate the average order amount for both groups and determine whether the discount coupon
leads to larger order sizes.

Answer:

Average Order Amount for Group A:


Average Order Amount = Total Order Amount / Number of Orders

= $10,000 / 500

= $20

Average Order Amount for Group B:

Average Order Amount = Total Order Amount / Number of Orders

= $12,500 / 600

= $20.83

Conclusion:

The average order amount for the group that received the discount coupon (Group B) is $20.83,
which is slightly higher than the average order amount for the group with no coupon (Group A) at
$20. This suggests that the discount coupon may lead to slightly larger order sizes.

Problem 4:

A fitness app wants to compare the effectiveness of two different workout routines on calorie
burning. They randomly assign users to two groups: Group A follows Routine X, and Group B
follows Routine Y. After a month of testing, the app collects the following data:

Group A:

- Total calories burned: 15,000

- Number of users: 200

Group B:

- Total calories burned: 16,500

- Number of users: 180

Calculate the average calories burned per user for both groups and determine which workout
routine is more effective.

Answer:
Average Calories Burned per User for Group A:

Average Calories Burned per User = Total Calories Burned / Number of Users

= 15,000 / 200

= 75

Average Calories Burned per User for Group B:

Average Calories Burned per User = Total Calories Burned / Number of Users

= 16,500 / 180

= 91.67

Conclusion:

The average calories burned per user for Group B (Routine Y) is 91.67, which is higher than the
average calories burned per user for Group A (Routine X) at 75. This suggests that Routine Y is more
effective in terms of calorie burning.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy