Categorical Data

In which of the following situations will you have to deal with categorical
data? (Please note that multiple options can be selected)

You want to know whether or not the page load time affects the
revenue.

✕ Incorrect
Feedback:
The data in this situation will be numeric.

You want to know whether or not a more personalised search

algorithm will have an impact on the conversion rate.

✓ Correct
Feedback:
The data in this situation will be categorical.

You want to know whether or not employee efficiency is related to

salary.

✕ Incorrect
Feedback:
The data in this situation will be numeric.

You want to check whether or not altering the flow of your checkout
funnel will lead to more purchases.

✓ CorrectYou missed this!

Feedback:
The data in this situation will be categorical.

Your answer is Wrong.

A/B Testing
What will your null hypothesis be?

Conversion rate with original funnel ≤ Conversion rate with modified

funnel

This study source was downloaded by 100000857563832 from CourseHero.com on 01-28-2023 00:32:46 GMT -06:00

https://www.coursehero.com/file/116646767/AB-Testingdocx/
 Conversion rate with original funnel = Conversion rate with modified
funnel

Conversion rate with original funnel ≠ Conversion rate with modified

funnel

Conversion rate with original funnel ≥ Conversion rate with modified

funnel

✓ Correct
Feedback:
As you want to test whether or not the modified funnel works better, your
null hypothesis should state that the original funnel has a higher conversion
rate than the modified funnel.

Your answer is Correct.

A/B Testing
In the control group (original funnel), 4,010 customers initiate a checkout, of
which 1,220 complete it. In the test group (modified funnel), 3,995
customers initiate a checkout, of which 1,110 complete it.

What is the p-value of the test? (Use the AB testing calculator to compute it.)

0.0047

0.0094

0.9953

✓ Correct
Feedback:
As you can see, the test group performed worse (a conversion rate of
27.78%) than the control group (a conversion rate of 30.42%). This indicates
evidence in favour of the null hypothesis. The computed p-value is 0.9953.

0.5214
Your answer is Correct.
A/B Testing
Which of the following statements is true?

This study source was downloaded by 100000857563832 from CourseHero.com on 01-28-2023 00:32:46 GMT -06:00

https://www.coursehero.com/file/116646767/AB-Testingdocx/
 You can reject the null hypothesis while testing at a 20% significance
level but not at 10%.

You can reject the null hypothesis while testing at a 10% significance
level but not at 5%.

You can reject the null hypothesis while testing at a 5% significance

level but not at 1%.

None of the above

✓ Correct
Feedback:
You can reject the null hypothesis if the ‘p-value < significance level’. In this
case, as the p-value is 0.9953, you cannot reject the null hypothesis at a
significance level of 1%, 5%, 10% or 20%.

Your answer is Correct.

Perform an A/B test
What will your null hypothesis be?

Quality of Design A ≥ Quality of Design B

Quality of Design A ≠ Quality of Design B

Quality of Design A ≤ Quality of Design B

Quality of Design A = Quality of Design B

✓ Correct
Feedback:
You want to test your default belief that both the designs are equally good.
Hence, this should be stated in the null hypothesis.

Your answer is Correct.

Perform an A/B test
What is the p-value of the test? (Use the AB testing calculator to compute it.)

This study source was downloaded by 100000857563832 from CourseHero.com on 01-28-2023 00:32:46 GMT -06:00

https://www.coursehero.com/file/116646767/AB-Testingdocx/
 0.1413

0.1477

✓ Correct
Feedback:
66.20% of people downloaded the e-book from the test group (Design B),
while 65.05% of people downloaded the e-book from the control group
(Design A). The computed p-value is 0.1477.

0.3072

0.5219
Your answer is Correct.
Perform an A/B test
Which of the following statements is true?

You can reject the null hypothesis while testing at a 20% significance
level but not at 10%.

✓ Correct
Feedback:
You can reject the null hypothesis if the ‘p-value < significance level’. In this
case, as the p-value is 0.1477, you can reject the null hypothesis at a
significance level of 20% but not at 1%, 5% or 10%.

You can reject the null hypothesis while testing at a 10% significance
level but not at 5%.

You can reject the null hypothesis while testing at a 5% significance

level but not at 1%.

You can reject the null hypothesis while testing at a 1% significance

level.
Your answer is Correct.

This study source was downloaded by 100000857563832 from CourseHero.com on 01-28-2023 00:32:46 GMT -06:00

https://www.coursehero.com/file/116646767/AB-Testingdocx/
 Graded Assessments
Note that the following questions are graded.
Perform an A/B test
What will your null hypothesis be?

Sales with Tagline A ≥ Sales with Tagline B

✓ Correct
Feedback:
You want to test whether Tagline B helps generate more sales than Tagline A.
Hence, your null hypothesis should reflect the initial assumption that Tagline
A helps generate more sales.

Sales with Tagline A = Sales with Tagline B

Sales with Tagline A ≤ Sales with Tagline B

Sales with Tagline A < Sales with Tagline B

Your answer is Correct.
Perform an A/B test
What is the p-value of the test? (Use the AB testing calculator to compute it.)

0.3614

0.1807

✓ Correct
Feedback:
6.01% of people bought the product from the test group (Tagline B), while
5.79% of people bought the product from the control group (Tagline A). The
computed p-value is 0.1807.

0.2775

This study source was downloaded by 100000857563832 from CourseHero.com on 01-28-2023 00:32:46 GMT -06:00

https://www.coursehero.com/file/116646767/AB-Testingdocx/
 0.0513
Your answer is Correct.
Perform an A/B test
Which of the following statements is true?

You can reject the null hypothesis while testing at a 20% significance
level but not at 10%.

✓ Correct
Feedback:
You can reject the null hypothesis if the ‘p-value < significance level’. In this
case, as the p-value is 0.1807, you can reject the null hypothesis at a
significance level of 20% but not at 1%, 5% or 10%.

You can reject the null hypothesis while testing at a 10% significance
level but not at 5%.

You can reject the null hypothesis while testing at a 5% significance

level but not at 1%.

You can reject the null hypothesis while testing at a 1% significance

level.
Your answer is Correct.

This study source was downloaded by 100000857563832 from CourseHero.com on 01-28-2023 00:32:46 GMT -06:00

https://www.coursehero.com/file/116646767/AB-Testingdocx/
Powered by TCPDF (www.tcpdf.org)