CBDA Free Practice Test

- (Topic 1)
A software company launched a new product in late 2016. The product manager is reviewing a Box and Whisker plot used to compare year-over-year sales, from 2017 to 2018. What is the conclusion he can make from this chart?

1. A. 2017 minimum and maximum sales are higher than 2018, and the 2017 median result is higher than the 2018 median result
2. B. 2017 minimum and maximum sales are higher than 2018, but the 2017 median result is lower than 2018 1st quartile result
3. C. 2018 minimum and maximum sales are higher than 2017, and the 2018 quartile results arehigher than 2017 quartile results
4. D. 2018 minimum and maximum sales are higher than 2017, and the 2018 1st quartile is higher than 2017 median result

Correct Answer: D

- (Topic 2)
A food and beverage company would like to administer a survey to obtain customer insights about a new cookie product recently launched. A data team is asked to build the survey paying careful attention to reduce the degree of sampling error.Which criteria would help the team meet this objective?

1. A. Large sample size and variation in the target population
2. B. Large sample size and random selection of the target population
3. C. Small sample size and specific subset of the target population
4. D. Small sample size and using customers who agreed to take the survey

Correct Answer: B
Sampling error is the difference between the results obtained from a sample and the results obtained from the population from which the sample is drawn1. Sampling error can affect the validity, reliability, and generalizability of the survey results2. To reduce the degree of sampling error, the data team should use a large sample size and a random selection of the target population. A large sample size means that the sample is more likely to represent the diversity and variability of the population, and that the results are more precise and accurate3. A random selection of the target population means that every member of the population has an equal chance of being included in the sample, and that the results are less biased and more representative4.
The other criteria would not help the team meet this objective, as they would increase the degree of sampling error. A large sample size and variation in the target population would not reduce the sampling error, as variation refers to the differences or heterogeneity within the population, not the sample. Variation in the target population can increase the sampling error, as it makes it harder to capture the true characteristics of the population with a sample5. A small sample size and specific subset of the target population would not reduce the sampling error, as they would make the sample less representative and more prone to bias. A small sample size means that the sample is less likely to reflect the diversity and variability of the population, and that the results are less precise and accurate. A specific subset of the target population means that the sample is not randomly selected, but based on some criteria or convenience, and that the results are more biased and less representative. A small sample size and using customers who agreed to take the survey would not reduce the sampling error, as they would also make the sample less representative and more prone to bias. A small sample size has the same drawbacks as mentioned above. Using customers who agreed to take the survey means that the sample is not randomly selected, but based on self-selection or voluntary response, and that the results are more biased and less representative.
References:1: Guide to Business Data Analytics, IIBA, 2020, p. 542: Introduction to Business Data Analytics: A Practitioner View, IIBA, 2019, p. 223: Data Analysis: The Definitive Guide, Tableau, 4: Data Analysis: The Definitive Guide, Tableau, 5: Data Analysis: The Definitive Guide, Tableau, . : Data Analysis: The Definitive Guide, Tableau, .
: Data Analysis: The Definitive Guide, Tableau, . : Data Analysis: The Definitive Guide, Tableau, . : Guide to Business Data Analytics, IIBA, 2020, p. 54. : Introduction to Business Data Analytics: A Practitioner View, IIBA, 2019, p. 22. : Data Analysis: The Definitive Guide, Tableau, . : Data Analysis: The Definitive Guide, Tableau, . : Data Analysis: The Definitive Guide, Tableau, . : Data Analysis: The Definitive Guide, Tableau, . : Data Analysis: The Definitive Guide, Tableau, . : Data Analysis: The Definitive Guide, Tableau, .

- (Topic 2)
A 3rd party is marketing an application for financial institutions to use for credit scoring. This application is an example of what type of analytics?

1. A. Descriptive analytics
2. B. Prescriptive analytics
3. C. Exploratory
4. D. Inferential

Correct Answer: B
Prescriptive analytics is the type of analytics that provides recommendations or suggestions for optimal actions or decisions based on data analysis. Prescriptive analytics uses techniques such as optimization, simulation, and decision analysis to generate and evaluate various scenarios and outcomes. Prescriptive analytics can help financial institutions to use credit scoring to determine the best loan offers, interest rates, and repayment terms for their customers, as well as to manage risk and compliance.
Prescriptive analytics is the most advanced and complex type of analytics, as it requires a high level of data quality, integration, and modeling, as well as human judgment and domain expertise. References:
✑ Certification in Business Data Analytics (IIBA ® - CBDA), IIBA, accessed on January 20, 2024.
✑ Business Data Analytics Certification - CBDA Competencies | IIBA®, IIBA, accessed on January 20, 2024.
✑ Guide to Business Data Analytics, IIBA, 2020, p. 15-16.

- (Topic 1)
As the team discusses how to utilize the results of their data analysis to put forth a business recommendation, an analyst on the team voices concern over the current organizational culture presenting a roadblock to their ability to influence business decision making. Which of the following would be a justifiable concern at this stage of the team's efforts?

1. A. Difficulty bringing business stakeholders to a shared understanding about value when sharing data assets across business domains
2. B. Changing the mindsets of business stakeholders to trust insights gleaned from data over experience and intuition
3. C. Applying a myopic view of data and establishing data silos which create roadblocks to exploring available data sources
4. D. Finding data that creates value creating difficulties, as not all data helps a business make better decisions

Correct Answer: B
A justifiable concern at this stage of the team??s efforts is changing the mindsets of business stakeholders to trust insights gleaned from data over experience and intuition. This is because some stakeholders may have a strong attachment to their own opinions or beliefs, and may resist or ignore data that contradicts them. This can create a barrier to data-driven decision making, which requires a culture of curiosity, openness, and evidence-based reasoning. The team needs to communicate the value and validity of their data analysis, and persuade the stakeholders to adopt a data-driven mindset12 References: 1: Use Data to Accelerate Your Business Strategy 2: Data-Driven Decision Making: A Step-by-Step Guide

- (Topic 1)
A professional association is funded by membership fees. The membership renewal occurs every 5 years. Although, they have a strong subscription rate each year, their renewal rate is low. They are working with an external firm specializing in Business Analytics to identify the groups of customers that have a high likelihood of cancelling their subscription after their first 5-year term ends. This type of study is called:

1. A. Untrained learning
2. B. Supervised learning
3. C. Trained learning
4. D. Unsupervised learning

Correct Answer: D
Unsupervised learning is a type of study that involves finding patterns or clusters in data without any predefined labels or outcomes. It is useful for exploring data and discovering hidden structures or groups of customers. For example, the professional association can use unsupervised learning to identify the characteristics of customers who are likely to cancel their subscription after their first 5-year term ends, and then design strategies to retain them12 References: 1: What is Unsupervised Learning? - IBM 2: Unsupervised Learning - IIBA BABOK Guide v3