IIBA IIBA Certifications CBDA Questions & Answers

• Question 41:

  The CustomerOrder entity will include information on all customer orders. Applying database normalization rules, which set of attributes will need to be normalized to avoid redundancies?

  1.

  Customerld

  2.

  CustomerPhone

  3.

  Orderld

  4.

  OrderDate

  5.

  ProductName

  6.

  ProductQuantity

  7.

  OrderTotal

  A. CustomerPhone

  B. ProductName

  C. ProductName ProductQuantity

  D. Orderld ProductName

  E. Customerld OrderDate

  Correct Answer: B

  Database normalization is the process of organizing the data in a database to reduce redundancy and improve integrity, consistency, and performance1. Database normalization rules are based on the concept of normal forms, which are levels of database design that meet certain criteria2. One of the most common normal forms is the third normal form (3NF), which states that a table should not have any transitive dependencies, meaning that a non-key attribute should not depend on another non-key attribute3. In the CustomerOrder entity, the set of attributes that will need to be normalized to avoid redundancies are ProductName and ProductQuantity, as they are non-key attributes that depend on another non-key attribute, Orderld. This means that the same product information may be repeated for different orders, which could lead to data inconsistency, duplication, or update anomalies. To normalize this set of attributes, a separate table should be created for the OrderDetails entity, which would have Orderld, ProductName, and ProductQuantity as its attributes, and Orderld and ProductName as its composite primary key.

  The other sets of attributes do not need to be normalized to avoid redundancies, as they do not violate the 3NF. CustomerPhone and ProductName are non-key attributes that depend on the primary key, Customerld and Orderld respectively, which is allowed by the 3NF. Orderld and ProductName are part of the composite primary key of the OrderDetails entity, which is also allowed by the 3NF. Customerld and OrderDate are both primary keys of the Customer and Order entities respectively, which are also allowed by the 3NF.

• Question 42:

  There were 7 students enrolled in the Introduction to Artificial Intelligence course. The scores from the final exam were as follows: 64, 70. 80, 80, 90, 98, 100 What is the mean and median for the outlined scores?

  A. 79.84.80

  B. 83.14,80

  C. 80,83.14

  D. 83.14.90

  Correct Answer: B

  The mean of a set of numbers is the sum of the numbers divided by the number of numbers. The median of a set of numbers is the middle value when the numbers are arranged in ascending or descending order. To find the mean and

  median of the given scores, we can use the following steps:

  To find the mean, we add up all the scores and divide by 7, the number of students. The mean is (64 + 70 + 80 + 80 + 90 + 98 + 100) / 7 = 582 / 7 = 83.14

  To find the median, we arrange the scores in ascending order: 64, 70, 80, 80, 90, 98, 100. Since there are an odd number of scores, the median is the middle score, which is 80.

  Therefore, the mean and median for the outlined scores are 83.14 and 80, respectively.

• Question 43:

  A job satisfaction study is being considered. Half of the employees of the company will be interviewed by senior managers and the other half of the employees will be interviewed by an external market research company, using the same set of questions. Which of the following might be a concern for using this approach to collect study data?

  A. Reliability

  B. Validity

  C. Timeliness

  D. Precision

  Correct Answer: A

  Reliability is the degree to which a data collection method produces consistent results under the same conditions1. In this case, the reliability of the study data might be compromised by the different interviewers (senior managers vs. external market research company), who might have different biases, expectations, or rapport with the employees. This could affect how the employees respond to the same set of questions, and thus introduce variability in the data. Validity, timeliness, and precision are not directly affected by the choice of interviewers, as they depend more on the quality, relevance, and accuracy of the questions and the data analysis.

• Question 44:

  A company wants to run a monthly promotion on batteries that cost 15 cents each and sells for 50 cents. At this price, they typically sell 1000 batteries and generate a profit of 35 cents per battery for a total profit of $350. The analytics team was asked to test two price points20% off (i.e. a sale price of 40 cents) and 40% off (i.e., a sale price of 30 cents). The survey data completed by 10000 participants was analyzed and showed that a 20% savings would result in sales of 1200 batteries and the 40% savings would result in 1800 batteries being sold. The team's initial recommendation was to recommend the 40% discount. Now that they are validating their recommendations, they decide to:

  A. Question why management would only want them to test two price points

  B. Change their recommendation realizing they have been victims of linear bias

  C. Redo the survey looking for a larger sample size

  D. Use their original recommendation given that the volume of sales is much higher

  Correct Answer: B

  Linear bias is a type of cognitive bias that assumes a linear relationship between two variables, when in fact the relationship may be more complex or nonlinear. In this case, the analytics team assumed that the higher the discount, the higher the sales and profit, without considering other factors that may affect customer behavior, such as price elasticity, perceived quality, or competition. By changing their recommendation, the team can avoid making a suboptimal decision that may result in lower profit or customer satisfaction.

• Question 45:

  Which attribute in the Customerissues entity would be categorized as unstructured data?

  1.

  CustomerlD

  2.

  ConcernCategory

  3.

  ConcernSubCategory

  4.

  AgentID

  5.

  ComplaintNotes

  6.

  IssueResolved(Y/N)

  A. ComplaintNotes

  B. ConcernCategory

  C. lssueResolved(Y/N)

  D. ConcernSubCategory

  Correct Answer: A

  Unstructured data is data that does not have a predefined format, structure, or schema, and that cannot be easily stored, processed, or analyzed by traditional databases or tools1. Unstructured data may include text, images, audio, video, or other types of data that are rich in information but complex and diverse in nature2. In the Customerissues entity, the ComplaintNotes attribute would be categorized as unstructured data, as it may contain free-form text that captures the details, sentiments, or emotions of the customers' complaints, and that may vary in length, language, tone, or style. The ComplaintNotes attribute would require special techniques, such as natural language processing, text mining, or sentiment analysis, to extract meaningful insights from the unstructured data3.

  The other attributes in the Customerissues entity would be categorized as structured data, as they have a predefined format, structure, or schema, and that can be easily stored, processed, or analyzed by traditional databases or tools4. Structured data may include numbers, dates, codes, categories, or other types of data that are simple and consistent in nature5. In the Customerissues entity, the CustomerlD, ConcernCategory, ConcernSubCategory, AgentID, and lssueResolved(Y/N) attributes would be categorized as structured data, as they may contain numeric, alphanumeric, or binary values that represent the identifiers, classifications, or statuses of the customers' issues, and that may have fixed lengths, ranges, or domains.

• Question 46:

  Allegra Consulting is planning on establishing an analytics system to track career progression of their consultants. Elicitation will be used to identify the required features. How would brainstorming be used to prepare for elicitation?

  A. To identify sources of business information to consider

  B. To identify the key metrics to be collected

  C. Determine the value for establishing the analytics system

  D. To choose the statistical methods required

  Correct Answer: A

  According to the Guide to Business Data Analytics, one of the tasks under the domain of "Identify the Research Questions" is to identify sources of business information to consider. This task involves reviewing existing business information, such as documents, reports, databases, and systems, to determine what data is available, relevant, and reliable for answering the research questions. This task also involves identifying any gaps or limitations in the existing information and proposing ways to address them.

• Question 47:

  A data scientist at a consumer goods company, has been asked to do a detailed analysis on customer profiles. The Data Scientist has identified an external data source that carries valuable additional information on their customers. The data scientist also identifies the address column as the most reliable column to join the internal data source with the external data source. Addresses may appear in different formats for example:

  File A = "13 Smith St"

  File B = "Unit 7, 13 Smith Street"

  Which of the following techniques would be useful in this situation?

  A. Deterministic linkage

  B. Probabilistic linkage

  C. Genetic linkage

  D. Cuff linkage

  Correct Answer: B

  Probabilistic linkage is a technique that uses statistical methods to match records from different data sources based on the similarity of key variables, such as name, address, date of birth, etc1. Probabilistic linkage can handle variations, errors, or missing values in the data, and assign a score or probability to each potential match2. Probabilistic linkage would be useful in this situation, as the address column may have different formats, spellings, or abbreviations in the internal and external data sources, and a deterministic linkage (which requires exact matches) might miss some valid matches or create false matches.

  Deterministic linkage is a technique that uses predefined rules or criteria to match records from different data sources based on the exact agreement of key variables, such as identifiers, codes, or hashes3. Deterministic linkage would not be useful in this situation, as the address column may not have consistent or unique values in the internal and external data sources, and a probabilistic linkage (which allows for some variation or uncertainty) might find more accurate matches or avoid false matches.

  Genetic linkage is a term used in genetics to describe the tendency of genes or DNA sequences that are located close together on a chromosome to be inherited together4. Genetic linkage is not relevant to this situation, as it has nothing to do with matching records from different data sources based on the address column.

  Cuff linkage is a term used in sewing to describe the process of attaching a cuff to a sleeve by stitching or fastening. Cuff linkage is not relevant to this situation, as it has nothing to do with matching records from different data sources based on the address column.

• Question 48:

  A movie production company wants to use analytics to decide which customers would choose to watch or not watch a particular movie after seeing a promotional teaser. The business analysis professional suggests they could make that prediction by identifying characteristics of the new movie and determining if the customer has watched other movies with similar characteristics.Thisis an example of using the following technique:

  A. Logistic regression

  B. Ouster analysis

  C. Integer programming

  D. Analysis of variance

  Correct Answer: A

  Logistic regression is a technique that can be used to model the probability of a binary outcome, such as choosing to watch or not watch a movie, based on one or more predictor variables, such as the characteristics of the movie and the customer's viewing history. Logistic regression can help the business analysis professional to identify the factors that influence the customer's decision and to estimate the likelihood of each customer's preference. Logistic regression can also be used to test hypotheses and to evaluate the performance of the predictive model.

• Question 49:

  An analyst is performing regression analysis and reviewing the results. They would like to rescale the variables in the model to more clearly reflect the relationship between the regression coefficients.Which technique could be used to rescale the variables?

  A. Dimension Reduction

  B. Mean Centering

  C. Normalization

  D. Clustering

  Correct Answer: C

  Normalization is a technique that rescales the values of the variables in a data set to a common range, such as [0,1] or [-1,1]. Normalization can help reduce the effect of outliers, improve the performance of some algorithms, and make the interpretation of the regression coefficients easier and more consistent. Normalization can be done using different methods, such as min-max scaling, z-score scaling, or unit vector scaling.

• Question 50:

  A fifty-year-old brick and mortar business is interested in determining the potential for selling their current products online. The sales director has asked the analytics team to predict future sales for their most popular product. A simple question is formed "Would you buy this product online?" The sales director would like to survey students from local colleges and universities within a 50km radius.As a result, the team will conclude:

  A. The sample size being considered may be too large to work with

  B. The research question will be easily answered with currently available data

  C. Focusing on a 50km radius will allow the team to complete the analysis quickly

  D. The survey will establish a poor study population

  Correct Answer: D

  According to the Guide to Business Data Analytics, a study population is the subset of the population that meets the eligibility criteria for the research question. A study population should be representative of the population of interest and relevant to the business problem or opportunity. In this situation, the survey will establish a poor study population because the students from local colleges and universities within a 50km radius may not reflect the characteristics, preferences, and behaviours of the potential online customers for the fifty-year-old brick and mortar business. The students may have different demographics, income levels, shopping habits, and needs than the target market for the business. Therefore, the survey results may not be generalizable or applicable to the population of interest and may not provide valid and reliable insights for predicting future sales.