- (Topic 1)
The analytics team is assessing the results of their analysis. They are surprised to find that their data indicates two events seem to be strongly related even though the general belief in the organization is that they are independent of each other. Knowing that this information will be used for decision making, they are concerned about presenting this data. At an impasse, the business analysis professional reminds them that the data can be presented as long as the team has:
Correct Answer:
D
The ability to rerun the data analysis and the results are the same is the condition that the team should have before presenting the data, because it is a technique that ensures the validity, reliability, and reproducibility of the data analysis. By rerunning the data analysis, the team can verify that the results are consistent and not affected by random errors, biases, or anomalies. The team can also confirm that the data analysis process is well- documented, transparent, and traceable, and that the results can be replicated by other analysts or stakeholders. This can minimize the risk of acting on the data, and increase the confidence and trust in the data analysis. References:
•Business Analysis Certification in Data Analytics, CBDA | IIBA®, CBDA Competencies, Domain 4: Interpret and Report Results
•Understanding the Guide to Business Data Analytics, page 9
•Business Data Analytics (IIBA®-CBDA Exam preparation) | Udemy, Section 4: Interpretand Report Results, Lecture 20: Data Validation and Verification
- (Topic 2)
Results of the data analysis have been analyzed and the team was confident with the results but also quite surprised the outcome was not what was expected. In pondering the value of what can be gleaned from the data, the team has no feasible solution to put forth to address the business need.A logical next step would be to:
Correct Answer:
A
According to the Guide to Business Data Analytics, the business analytics cycle is an iterative process that consists of four phases: identify the research questions, source data, analyze data, and interpretand report results. The cycle can be repeated as many times as needed until the business problem or opportunity is addressed or resolved. In this situation, the team was confident with the results but also surprised that the outcome was not what was expected. This means that the initial research question may not have been relevant, specific, or testable enough to provide a feasible solution for the business need. Therefore, a logical next step would be to repeat the business analytics cycle with the formation of a new research question that is more aligned with the business goal, scope, and context.
References: Guide to Business Data Analytics, page 47-48; CBDA Exam Blueprint, page 7; [Introduction to Business Data Analytics: A Practitioner View], page 15.
- (Topic 2)
A data dictionary is being developed for a dataset describing a company's customer base. Within the data dictionary, which of the following represents a composite data element?
Correct Answer:
A
A composite data element is a data element that is made up of smaller units called sub-elements, which are separated by a sub-element separator character, such as a colon (:). For example, ITEMNO is a composite data element that consists of three sub- elements: part number, aisle number, and bin number. A street address is also a composite data element that can consist of sub-elements such as street number, street name, city, state, and zip code. First name, total sale, and birthdate are simple data elements that do not have sub-elements.
References:Data Elements - IBM, UN/EDIFACT Syntax Rules
- (Topic 1)
The analytics team has been asked to determine if the organization should launch their highest revenue generating product into the North American market. To date, this has only been available in Eastern Europe. To answer this, the team formulates several research questions, including:
Correct Answer:
D
One of the steps in identifying the research questions for business data analytics is to assess the feasibility and desirability of the proposed solution or change1. This involves understanding the needs, preferences, and satisfaction of the existing and potential customers. Therefore, asking whether the existing customers really like the product is a relevant research question for the analytics team. References: 1: Guide to Business Data Analytics, IIBA, 2020, p. 22.
- (Topic 1)
A research marketer is interested in collecting information about the spending habits of families in North America. Concerned about the volume of data required to conduct the research, they choose to use sampling. The dataset is sourced using all credit card transactions from a leading North American credit card company for Quarter 1 of the prior year. The sample used is:
Correct Answer:
D
The sample used in this case is biased, meaning that it is not representative of the population of interest. The population of interest is the families in North America, but the sample is drawn from only one source of data: the credit card transactions from a leading North American credit card company. This sample excludes the families who do not use credit cards, or who use other credit card companies, or who use other payment methods. Therefore, the sample is not random or fair, and it may introduce sampling bias into the research results12 References: 1: Sampling Methods | Types, Techniques & Examples 2: Sampling Bias - an overview | ScienceDirect Topics
