IIBA IIBA Certifications CBDA Questions & Answers

• Question 131:

  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:

  A. Statistically representative

  B. Not relevant

  C. Too large to be helpful

  D. Biased

  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 results

• Question 132:

  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:

  A. Review the results with management ahead of time and highlight any potential risk of using this data

  B. Confidence that the correlation will reliably occur in the future and the risk of acting on this is low

  C. Followed all rules for data analysis endorsed as organizational standards so the risk of acting on this is low

  D. The ability to rerun the data analysis and the results are the same thereby minimizing the risk of acting on this

  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.

• Question 133:

  What is the relationship between a Customer entity and an Order entity, where a customer entry will be present in the Customer entity only if they have made an order?

  A. one-to-many

  B. many-to-many

  C. one-to-one

  D. zero-to-one

  Correct Answer: D

  The relationship between a Customer entity and an Order entity, where a customer entry will be present in the Customer entity only if they have made an order, is a zero-to-one relationship. This means that for each record in the Order entity, there can be either zero or one record in the Customer entity that is related to it. This implies that the Order entity is optional for the Customer entity, and the Customer entity is mandatory for the Order entity12

• Question 134:

  An analyst is looking at a particular dataset that includes the scores across all 8th grade students, across three schools. The analyst is trying to determine which type of statistics average to use to best represent the results. On looking through the dataset, the analyst has identified a few extreme outliers. As a result, the analyst was led to use the following type of average:

  A. Median

  B. Range

  C. Mean

  D. Mode

  Correct Answer: A

  The median is the type of statistics average that the analyst should use to best represent the results, because it is a measure of central tendency that divides the data set into two equal halves. The median is the middle value of the data set when it is arranged in ascending or descending order. The median is not affected by extreme outliers, unlike the mean, which is the arithmetic average of the data set. The median can give a more accurate representation of the typical score of the 8th grade students across the three schools. Options B, C, and D are not types of statistics average, but types of statistics measures that describe other aspects of the data set. The range is a measure of dispersion that shows the difference between the highest and the lowest values of the data set. The mean is a measure of central tendency that shows the sum of the values of the data set divided by the number of values. The mode is a measure of central tendency that shows the most frequent value of the data set.

• Question 135:

  A company wants to gauge the thoughts of their employees towards a new company product. On the 25th of March the interviewer makes a list of all employees who were at work on that day and then chooses a subset of those employees to interview. Which term describes the list of all employees present on March 25th?

  A. Population of interest

  B. Survey sample

  C. Sampling frame

  D. Sample weights

  Correct Answer: C

  The sampling frame is the term that describes the list of all employees present on March 25th, because it is a technique that defines the set of elements from which a sample is drawn. The sampling frame should ideally match the population of interest, which is the group of elements that the researcher wants to study or make inferences about. In this case, the population of interest is the employees of the company, and the sampling frame is the subset of employees who were at work on a specific day. The survey sample is the technique that selects a portion of the sampling frame to participate in the survey. The sample weights are the technique that assigns different values or importance to each element in the sample, based on their representation in the population.

• Question 136:

  The analytics team has been asked to assess sales data from their company's website with the hopes of providing insights to help increase online sales. It's the first time the team is looking at this specific data and they are concerned about the quality of data that has been captured. They decide to use the following approach as the next step:

  A. Trend Analysis

  B. Classification analysis

  C. Data Analysis

  D. Exploratory analysis

  Correct Answer: D

  Exploratory analysis is the approach that the analytics team should use as the next step, because it is a technique that allows them to examine the quality, structure, and characteristics of the data, without making any assumptions or hypotheses. Exploratory analysis can help the team identify any issues or anomalies in the data, such as missing values, outliers, or errors, and decide how to handle them. Exploratory analysis can also help the team discover any patterns, trends, or relationships in the data, and generate new research questions or hypotheses for further analysis.

• Question 137:

  An analyst has just completed building a data model that shows the table structures including table names, table relationships with primary and foreign keys and column names with respective data types. What type of data model has the analyst just built?

  A. Physical

  B. Hierarchical

  C. Conceptual

  D. Logical

  Correct Answer: A

  A physical data model is the most detailed and specific type of data model, which shows how the data is stored, accessed, and manipulated in the database. It includes the table structures, column names, data types, primary and foreign keys, constraints, indexes, and other physical attributes of the data

• Question 138:

  The research question prompting the use of analytics is well-defined. The team obtains the results and determines that the source data did not provide reliable results. As a result of this finding, the team modifies the original question to one that can be answered by the data. What is a risk that could impact the value of this analysis?

  A. The objective of the original research may not be met

  B. Timelines will be pushed out making stakeholders unhappy

  C. Increased costs associated with the source data

  D. The quality of the analysis may be negatively impacted

  Correct Answer: A

  The risk that could impact the value of this analysis is that the objective of the original research may not be met, because the team modified the research question to fit the data, rather than finding the data that fits the research question. This could lead to a loss of alignment between the research question and the business problem, stakeholder needs, or analytical methods. The team may end up answering a different or less relevant question than the one they intended to answer, and thus provide less valuable insights or recommendations.

• Question 139:

  A new dataset describing employee salaries is received by a company. A colleague wonders whether a variable follows a Gaussian distribution. Which of the following plots would demonstrate this?

  A. Normal probability plot

  B. Scatterplot

  C. Boxplot

  D. Lowess curve

  Correct Answer: A

  A normal probability plot is a graphical technique that can be used to check if a variable follows a Gaussian distribution. It plots the observed values of the variable against the expected values under the normal distribution. If the variable is normally distributed, the points should form a straight line. A scatterplot, a boxplot, and a lowess curve are not suitable for testing normality, as they do not compare the observed values with the theoretical values of the normal distribution.

• Question 140:

  A large car manufacturer is interested in comparing the number of sales for a specific model of electric car across all 50 US states.

  The data analytics team sourced and acquired the data, and the business analyst created the model to compare sales across states.

  In a meeting to review the results, the feedback received included several complaints concerning an inability to distinguish the number of sales per state. What model would result in such confusion?

  A. Bullet chart

  B. Dual axis chart

  C. Bar chart

  D. Pie chart

  Correct Answer: D

  A pie chart is a circular chart that shows the proportion of each category in a whole by dividing the circle into slices. A pie chart would result in confusion when comparing the number of sales for a specific model of electric car across all 50 US states, because it is difficult to compare the angles and areas of the slices, especially when there are many categories with similar values. A pie chart also does not show the absolute values of each category, unless they are labeled or annotated. A better alternative would be a bar chart, which can show the number of sales for each state along a common axis, making it easier to compare and rank the values