Knowledge + Practice

CompTIA Data+ (DA0-002) (DA0-002) — Questions 226–300

1004 questions total · 14pages · All types, answers revealed

Take a mock exam Exam hub

Page 4 of 14

226

Multi-Selectmedium

A data analyst wants to visualize the distribution of customer ages, including quartiles and potential outliers, for a dataset with 10,000 records. Which TWO chart types are appropriate? (Choose two.)

Select 2 answers

A.Box plot

B.Pie chart

C.Line chart

D.Treemap

E.Histogram

AnswersA, E

A box plot displays the median, quartiles and whiskers, and plots points beyond 1.5×IQR as outliers. This directly satisfies the requirement to show distribution, quartiles and potential outliers across 10,000 age records without plotting every value.

Why this answer

A box plot (A) is correct because it directly displays the five-number summary — minimum, first quartile (Q1), median, third quartile (Q3), and maximum — along with potential outliers plotted beyond the 1.5×IQR fences, which is exactly what the analyst needs for quartiles and outliers. A histogram (E) is also correct because it bins the 10,000 age values into intervals and shows the frequency distribution, revealing shape, skew, and modality of the ages. A pie chart (B) is unsuitable because it shows parts-of-a-whole proportions for categorical data, not a numeric distribution.

A line chart (C) is inappropriate since it implies a continuous sequence or trend over an ordered variable such as time, not a frequency distribution. A treemap (D) displays hierarchical part-to-whole relationships using nested rectangles, which does not convey quartiles or outliers.

Exam trap

DA0-002 often tests whether candidates confuse charts for categorical part-to-whole data (pie, treemap) with charts for continuous distribution analysis (box plot, histogram).

Full explanation →

227

MCQhard

A data analyst needs to combine customer data from a MySQL transactional database with product data from a MongoDB document store to create a unified view for reporting. The analyst uses a SQL query that joins the tables after extracting data from both sources. Which database concept is being applied?

A.View

B.Join

C.Stored procedure

D.Index

AnswerB

Joins combine rows from two or more tables based on a related column.

Why this answer

(Join) because the scenario describes combining data from two different sources—MySQL and MongoDB—into a unified view using a SQL query that joins the tables after extraction. This is a classic example of a cross-source join, where data from disparate databases is merged based on a common key, which is the fundamental purpose of a JOIN operation in SQL.

Exam trap

The trap here is that candidates may confuse a 'view' with a cross-source join, thinking a view can span multiple databases, but a view is limited to a single database and cannot directly reference tables from different database systems like MySQL and MongoDB.

How to eliminate wrong answers

Option A (View) is wrong because a view is a saved SQL query that presents data from one or more tables within the same database, not a mechanism to combine data from different source systems like MySQL and MongoDB. Option C (Stored procedure) is wrong because a stored procedure is a precompiled collection of SQL statements that performs a specific task within a single database, not a concept for joining data across heterogeneous data stores. Option D (Index) is wrong because an index is a data structure that improves the speed of data retrieval operations on a table, not a method for combining data from multiple sources.

Full explanation →

228

Multi-Selectmedium

A data analyst is designing a self-service reporting platform for business users. Which TWO practices will help ensure data consistency and trust? (Select TWO.)

Select 2 answers

A.Implement a single version of truth

B.Allow users to create their own data sources

C.Grant row-level security to all users

D.Remove all data lineage tracking

E.Provide a data dictionary for all metrics

AnswersA, E

A single version of truth consolidates metrics and definitions into one governed dataset, eliminating conflicting figures across reports. This directly satisfies the consistency and trust requirement, since business users draw from identical, curated data rather than divergent extracts.

Why this answer

Option A (Implement a single version of truth) is correct because consolidating reporting on one governed, authoritative dataset (e.g., a curated gold layer or semantic model) eliminates conflicting figures that arise when users pull from multiple duplicated sources, directly ensuring consistency and trust. Option E (Provide a data dictionary for all metrics) is correct because documenting each metric's exact definition, calculation, source column, and owner gives business users a shared, unambiguous meaning for every KPI, preventing misinterpretation and reinforcing confidence in the numbers. Options B, C, and D do not belong: letting users create their own data sources (B) proliferates ungoverned, inconsistent datasets; granting row-level security to all users (C) is an access-control measure that does not by itself guarantee metric consistency; and removing data lineage tracking (D) destroys the auditability and traceability that underpin trust in the data.

Full explanation →

229

MCQmedium

A financial analyst needs to show how net income is derived from revenue, costs, and expenses over a period. The chart should highlight the contribution of each component. Which chart type is most appropriate?

A.Line chart

B.Area chart

C.Waterfall chart

D.Pie chart

AnswerC

A waterfall chart models cumulative addition and subtraction, starting at revenue and stepping through costs and expenses to arrive at net income. Each bar isolates one component's contribution, directly satisfying the requirement to show how net income is derived and to highlight each component's individual effect.

Why this answer

A waterfall chart shows how an initial value is affected by a series of positive and negative changes, making it ideal for financial breakdowns.

Full explanation →

230

MCQmedium

Which database concept ensures that data in one table corresponds to data in another table, preventing orphan records?

A.Index

B.Referential integrity

C.Foreign key

D.Primary key

AnswerB

Referential integrity enforces foreign key constraints so every value in a child table matches an existing primary key in the parent table, directly preventing orphan records. This satisfies the stem's requirement that data in one table corresponds to data in another, which entity integrity and normalisation do not guarantee.

Why this answer

Referential integrity is the database principle that ensures relationships between tables remain consistent — every foreign key value must match an existing primary key value, preventing orphan records. It is enforced through constraints like FOREIGN KEY with ON DELETE/UPDATE actions.

Exam trap

DA0-002 often tests the distinction between the foreign key (the column/constraint) and referential integrity (the concept/rule), causing candidates to select the implementation rather than the principle.

How to eliminate wrong answers

Option A is wrong because an index is a performance structure for faster lookups, not a consistency mechanism. Option C is wrong because a foreign key is the column-level construct that implements referential integrity, but the concept itself is referential integrity — the question asks for the concept. Option D is wrong because a primary key uniquely identifies rows in its own table and does not by itself prevent orphans in a related table.

Full explanation →

231

MCQmedium

A data analyst creates a dashboard showing average order value by region. The chart indicates that one region has an unusually high average. Investigation reveals that the region has very few orders, but one large purchase inflates the average. Which data transformation should the analyst apply to improve the visualization?

A.Apply logarithmic scaling to the y-axis

B.Use median instead of mean for aggregation

C.Change the chart type to a pie chart

D.Remove all outliers from the dataset

AnswerB

The mean is skewed by that single large purchase because the region has few orders, so it no longer represents typical order value. The median resists outliers, giving a central value that reflects the region's actual orders and corrects the inflated chart.

Why this answer

The mean is highly sensitive to extreme values, so a single large purchase in a region with few orders skews the average upward and misrepresents the typical order value. The median is a robust measure of central tendency that resists outlier influence, so switching the aggregation from mean to median gives a truer picture of typical order value per region.

Exam trap

The trap is that candidates focus on visual fixes (log scale, chart type) when the real issue is the statistical aggregation; the exam tests whether you recognize mean vs. median robustness to outliers.

How to eliminate wrong answers

Option A is wrong because logarithmic scaling compresses the y-axis visually but does not change the underlying skewed statistic — the inflated mean is still plotted. Option C is wrong because changing to a pie chart does not address the aggregation problem and pie charts are poor for comparing averages across regions. Option D is wrong because removing all outliers discards legitimate business data (a real large purchase) and biases the analysis rather than choosing a robust statistic.

Full explanation →

232

Multi-Selecteasy

Which TWO are common pitfalls when communicating data insights?

Select 2 answers

A.Explaining assumptions clearly

B.Using misleading scales on charts

C.Including a clear call to action

D.Providing context for the data

E.Overloading the audience with too many visuals

AnswersB, E

Truncated or non-zero-baseline axes exaggerate small differences, so audiences draw conclusions the underlying data does not support. This satisfies the pitfall constraint because the distortion is introduced by presentation choices rather than the analysis itself, misleading decision-makers.

Why this answer

Option B is correct because using misleading scales on charts—such as truncating the y-axis so it does not start at zero or using inconsistent intervals—distorts the visual representation of data and can lead the audience to draw false conclusions, making it a classic data-communication pitfall. Option E is correct because overloading the audience with too many visuals (chart junk or excessive dashboards) exceeds cognitive capacity, dilutes the key message, and causes the audience to miss the most important insights. By contrast, option A (explaining assumptions clearly) is a best practice that builds trust and transparency, not a pitfall.

Option C (including a clear call to action) is also a best practice because it tells the audience what to do with the insight. Option D (providing context for the data) is likewise a best practice, since context such as baselines, benchmarks, and timeframes is essential for correct interpretation.

Exam trap

CompTIA often tests the distinction between best practices and common pitfalls, so the trap here is that candidates may confuse beneficial actions (like explaining assumptions or providing context) with pitfalls, leading them to select those as wrong answers instead of recognizing them as correct practices.

Full explanation →

233

MCQhard

A data analyst is designing a dashboard that will be viewed on both desktop monitors and mobile devices. The dashboard contains several charts and KPI cards. Which design approach best ensures usability across these devices?

A.Use small fonts and compact charts to fit more information on mobile screens

B.Use a fixed-width layout with horizontal scrolling for mobile users

C.Create two separate dashboards: one for desktop and one for mobile

D.Implement a responsive layout that rearranges and resizes elements based on screen size

AnswerD

A responsive layout automatically adjusts the arrangement and size of dashboard elements to fit the screen, ensuring that charts and KPIs remain readable and accessible on any device. This approach eliminates horizontal scrolling and optimizes the viewing experience for both desktop and mobile users.

Why this answer

Implementing a responsive layout is the best approach because it dynamically adjusts the dashboard's elements to fit any screen size, ensuring that charts and KPIs are legible and interactive on both desktop and mobile. This avoids the pitfalls of horizontal scrolling, separate maintenance, and unreadable compact designs.

Exam trap

The trap here is thinking that shrinking everything to fit mobile is acceptable, but that sacrifices readability; responsive design is about reflowing content, not just scaling it down.

Full explanation →

234

MCQmedium

A data analyst is pulling data from a production database for a report. The database contains customer orders with a column 'order_date'. The analyst notices that some orders have dates in the future. Which data quality issue does this represent?

A.Invalid data type

B.Inconsistent data

C.Missing data

D.Violation of business rules

AnswerD

Future orders are not valid per business rules, indicating a data quality issue.

Why this answer

Future order dates violate a business rule that order_date must be in the past or present. This is a classic data integrity issue where the data does not conform to domain-specific constraints, such as 'order_date <= CURRENT_DATE'. The analyst should flag this as a violation of business rules, not a data type or consistency problem.

Exam trap

The trap here is that candidates confuse 'invalid data type' (Option A) with 'invalid data value' — the data is of the correct type but violates a logical business rule, which is a distinct quality issue often tested in DA0-001.

How to eliminate wrong answers

Option A is wrong because the column 'order_date' is of a valid date data type (e.g., DATE or TIMESTAMP), so there is no data type mismatch. Option B is wrong because inconsistent data refers to contradictory values across related columns (e.g., different date formats), not a single column containing future dates. Option C is wrong because missing data would involve NULL or empty values, not dates that are present but invalid according to business logic.

Full explanation →

235

MCQeasy

A junior data analyst at an e-commerce company wants to query a production customer table that contains email addresses and purchase history. Company policy states that analysts may access only the columns needed for their assigned task and must request elevated access for sensitive fields. Which data governance principle is the policy enforcing?

A.Data sovereignty

B.Data archiving

C.Data redundancy

D.Least privilege

AnswerD

Least privilege means granting users only the minimum access necessary to perform their assigned tasks. Allowing the analyst to query only needed columns and requiring an elevated request for sensitive fields such as email addresses directly implements this principle, making it the correct governance concept in this scenario.

Why this answer

Least privilege restricts access to only what is required for a user's role and task. The policy lets the junior analyst query necessary columns while requiring an elevated request for sensitive fields like email addresses, which is a direct application of that principle. It reduces exposure of personal data without blocking legitimate analytical work.

Exam trap

The trap here is confusing least privilege with data minimization, since both limit data, but least privilege governs user access rights while minimization governs collection and retention.

Full explanation →

236

MCQmedium

A data analyst at a hospital is building a report on patient readmissions. The analyst needs to combine the patient demographics table (which uses PatientID as the primary key) with the admissions table (which uses AdmissionID as the primary key and also contains PatientID as a foreign key). Which type of operation should the analyst use to combine these two tables?

A.Aggregate

B.Join

C.Intersect

D.Union

AnswerB

A join combines columns from two or more tables based on a related column, such as PatientID. Here, the demographics and admissions tables share PatientID, so a join will correctly merge patient details with admission records. This is the standard relational operation for combining tables horizontally when a common key exists.

Why this answer

Combining two tables that share a common key, such as PatientID, requires a join operation. A join matches rows from both tables on the key and returns columns from both, enabling the analyst to see patient demographics alongside each admission. Other set operations like union or intersect do not align columns properly for this purpose.

Exam trap

The trap here is confusing set operations like union or intersect with join operations, which are used for different purposes in relational data combination.

Full explanation →

237

MCQhard

A data analyst is building a report that will be distributed as a PDF to stakeholders. The report includes a bar chart showing sales by region. The analyst wants to ensure that the chart is accessible to color-blind readers. Which action should the analyst take?

A.Use a red-green color scheme to highlight high and low performers

B.Convert the bar chart to a pie chart with a legend

C.Use a monochromatic color scheme with varying shades of blue

D.Add data labels to each bar and use a color palette that is distinguishable for common types of color blindness

AnswerD

Adding data labels ensures that the values are directly readable regardless of color perception. Pairing this with a colorblind-friendly palette, such as the Okabe-Ito palette, which avoids problematic color combinations, makes the chart accessible to readers with various color vision deficiencies.

Why this answer

To make a bar chart accessible for color-blind readers, the analyst should add data labels and use a colorblind-friendly palette. Data labels provide direct value readouts, while a carefully chosen palette ensures that colors are distinguishable for common types of color vision deficiency. This combination guarantees that the chart is interpretable by all stakeholders.

Exam trap

The trap here is assuming that any color scheme works as long as it looks distinct to the designer, but color blindness affects perception, so relying solely on color without labels or patterns is risky.

Full explanation →

238

Multi-Selectmedium

A data analyst is preparing a dataset for a machine learning algorithm that assumes normally distributed features. Which TWO data transformation methods should the analyst consider to achieve this?

Select 2 answers

A.Square root transformation

B.Log transformation

C.One-hot encoding

D.Z-score standardization

E.Min-max normalization

AnswersA, B

The square root transformation compresses right-skewed data, reducing the influence of large values and pulling the distribution toward normality. It satisfies the algorithm's normality assumption for moderately skewed features, particularly count data. Unlike logarithms, it handles zero values and is milder, making it suitable when skew is moderate rather than severe.

Why this answer

The square root transformation (A) is correct because it is a variance-stabilizing transformation that compresses right-skewed data and can make moderately skewed distributions more symmetric and closer to normal, which suits algorithms assuming normally distributed features. The log transformation (B) is also correct because it strongly reduces right skew and pulls in large outliers, making positively skewed data (e.g., exponential or multiplicative data) approximate a normal distribution. One-hot encoding (C) is not appropriate here because it converts categorical variables into binary indicator columns and does not change the distribution shape of numeric features toward normality.

Z-score standardization (D) only rescales features to mean 0 and standard deviation 1 without altering skewness or the underlying distribution shape. Min-max normalization (E) merely rescales values to a fixed range such as [0,1] and likewise does not make a non-normal distribution normal.

Exam trap

The trap here is confusing transformations that change distribution shape (e.g., square root, log) with those that only rescale (e.g., standardization, normalization), leading candidates to select scaling methods when the goal is to achieve normality.

Full explanation →

239

MCQmedium

A retail company with 500 stores across North America wants to visualize its sales performance. The dataset includes store ID, region (Northeast, Southeast, Midwest, West), product category (Electronics, Clothing, Home Goods), monthly sales (in dollars), and date (from January 2018 to December 2023). The data has missing values for about 5% of store-month combinations, and a few stores have reported sales that are 10 times higher than the average for their region due to grand opening events. The goal is to create a dashboard that shows monthly sales trends for each region and product category, and allows users to identify which categories are driving growth. Which approach should the analyst take?

A.Use a stacked bar chart showing total sales by month, with each bar segmented by region and category

B.Create a line chart with month on the x-axis, sales on the y-axis, and separate lines for each region and category; check for outliers and consider annotating them

C.Create a scatter plot of sales vs. month with dots colored by region

D.Remove all stores with outlier sales and then create a line chart of the cleansed data

AnswerB

A line chart with month on the x-axis, sales on the y-axis, and separate lines per region and category directly satisfies the requirement to show monthly trends and reveal which categories drive growth. Checking and annotating the grand-opening outliers prevents those 10x spikes distorting the trend lines.

Why this answer

A line chart with month on the x-axis and separate lines for each region and category best shows monthly sales trends over time and allows comparison across categories to identify growth drivers. Annotating outliers (grand opening spikes) preserves data integrity while explaining anomalies, which is critical since removing them would distort the trend analysis.

Exam trap

The trap is choosing to remove outliers (Option D) as a 'data cleaning' step, when in fact legitimate business events like grand openings should be annotated, not deleted, to preserve analytical accuracy.

How to eliminate wrong answers

Option A is wrong because a stacked bar chart obscures individual category trends and makes it hard to compare growth rates across regions and categories over time. Option C is wrong because a scatter plot of sales vs. month does not effectively show trends or category-level growth and becomes cluttered with 500 stores. Option D is wrong because removing outlier stores entirely discards legitimate grand-opening data and biases the analysis, rather than annotating or handling them appropriately.

Full explanation →

240

MCQmedium

A retail analytics team is building a monthly dashboard that must show the running total of sales for each store from the first day of the fiscal year through the current transaction date. The sales table contains one row per transaction with columns store_id, transaction_date, and sale_amount. The analyst needs a query that returns every transaction row with an additional column showing the cumulative sales for that store up to and including that transaction. Which SQL approach should the analyst use?

A.Use a SUM(sale_amount) OVER (PARTITION BY store_id ORDER BY transaction_date) window expression.

B.Use a correlated subquery with SUM(sale_amount) and a WHERE clause that filters transaction_date to the current month only.

C.Use GROUP BY store_id with SUM(sale_amount) and a HAVING clause on transaction_date.

D.Use a self-join on store_id where the joined transaction_date is greater than or equal to the outer transaction_date, then SUM the inner sale_amount.

AnswerA

This window expression computes a running total because the ORDER BY inside OVER defines the cumulative frame from the start of the partition to the current row, and PARTITION BY store_id restarts the accumulation for each store. It preserves every transaction row while adding the cumulative value, which matches the requirement exactly.

Why this answer

A window function with PARTITION BY store_id and ORDER BY transaction_date produces a running total per store while keeping every transaction row. The other choices either collapse rows through aggregation, reverse the accumulation logic, or restrict the date range so the total is not year-to-date. The window approach is the standard, efficient way to add a cumulative column without losing detail.

Exam trap

The trap here is assuming any SUM with GROUP BY yields a running total, when grouping actually collapses rows and cannot produce a per-row cumulative value.

Full explanation →

241

MCQmedium

A dataset contains employee salaries ranging from $30,000 to $200,000. An analyst wants to scale the salaries to a range of 0 to 1 for use in a distance-based clustering algorithm. Which method should they use?

A.Log transformation

B.Robust scaling

C.Min-max normalization

D.Z-score standardization

AnswerC

Min-max normalization rescales each value using (x − min)/(max − min), mapping the $30,000–$200,000 salary range linearly onto 0–1. This preserves relative distances, which distance-based clustering requires, unlike z-score standardisation, which centres on the mean with unbounded output.

Why this answer

Min-max normalization rescales values linearly to a fixed range, typically [0, 1], using the formula (x − min) / (max − min). This is exactly what the analyst needs for a distance-based clustering algorithm, where features on different scales would otherwise dominate the distance metric. It preserves the relative ordering and shape of the distribution while bounding all values between 0 and 1.

Exam trap

DA0-002 often tests normalization vs. standardization — candidates pick z-score because it is commonly used, missing that the question explicitly requires a 0-to-1 bounded range that only min-max normalization provides.

How to eliminate wrong answers

Option A is wrong because a log transformation compresses skewed data but does not bound values to [0, 1], so it fails the stated range requirement. Option B is wrong because robust scaling uses the median and IQR, centering data around zero with no fixed upper bound — it handles outliers but does not produce a 0–1 range. Option D is wrong because z-score standardization produces a mean of 0 and standard deviation of 1, yielding values that can be negative or exceed 1, not a bounded [0, 1] range.

Full explanation →

242

MCQhard

In a table 'employee_hierarchy' with columns 'employee_id', 'manager_id', and 'employee_name', an analyst needs to generate a list of all employees under a specific manager, including multiple levels of subordinates. Which SQL construct is most appropriate for querying this hierarchical data efficiently?

A.Recursive CTE

B.Window function with PARTITION BY

C.Subquery in WHERE clause

D.Self-JOIN with WHERE clause

AnswerA

A recursive CTE satisfies the multi-level requirement by self-referencing: the anchor member selects direct subordinates where manager_id matches the given manager, then the recursive member joins employee_hierarchy back to the CTE on manager_id = employee_id, iterating until no rows return. This traverses the full hierarchy in one statement, unlike fixed self-joins.

Why this answer

Recursive CTEs are designed to handle hierarchical data by repeatedly joining a CTE to itself until all levels are included.

Full explanation →

243

MCQmedium

A healthcare organization acquires data from multiple hospitals with different patient record systems. The data includes patient IDs but no common identifier across systems. Which technique should be used to link records?

A.Merge all records without deduplication

B.Generate random unique IDs for each system

C.Manually match records for all patients

D.Probabilistic record linkage using name, DOB, and ZIP

AnswerD

Without a common identifier, deterministic joins fail, so probabilistic record linkage compares name, date of birth and ZIP to compute match weights across hospitals. This satisfies the stem's constraint of linking records lacking a shared key, tolerating minor discrepancies between systems.

Why this answer

Probabilistic record linkage using name, DOB, and ZIP is the appropriate technique when there is no common identifier. It calculates the likelihood that two records refer to the same entity based on matching fields, accounting for errors and variations. This is standard in healthcare data integration.

Exam trap

DA0-002 often tests the misconception that deterministic matching (e.g., exact ID) is always possible, ignoring the need for probabilistic methods when identifiers are absent.

How to eliminate wrong answers

Option A is wrong because merging without deduplication would create duplicate and inconsistent records, leading to data quality issues. Option B is wrong because generating random unique IDs for each system would not link records across systems; it would only create new identifiers. Option C is wrong because manually matching records for all patients is impractical and error-prone for large datasets.

Full explanation →

244

MCQeasy

Which SQL aggregate function would an analyst use to calculate the average value of a numeric column?

A.SUM

B.AVG

C.COUNT

D.MEDIAN

AnswerB

AVG computes the arithmetic mean of a numeric column, directly satisfying the requirement to calculate an average value. Unlike SUM, which totals values, or COUNT, which tallies rows, AVG divides the sum by the count of non-NULL entries, returning the precise average the analyst needs.

Why this answer

The AVG aggregate function computes the arithmetic mean of a numeric column, which is exactly what the analyst needs. It ignores NULL values by default and returns a single average value for the specified column or expression.

Exam trap

DA0-002 often tests whether candidates confuse AVG with SUM or COUNT, or assume MEDIAN is a standard SQL aggregate, leading them to select a function that returns a total or a count instead of a mean.

How to eliminate wrong answers

Option A is wrong because SUM adds all values together and returns a total, not an average. Option C is wrong because COUNT returns the number of rows or non-null values, not a numeric average. Option D is wrong because MEDIAN is not a standard SQL aggregate function in most database platforms and returns the middle value, not the mean.

Full explanation →

245

MCQeasy

An organization wants to assign responsibility for data quality and metadata management. Which role is primarily accountable for defining data standards and ensuring data quality across a specific domain?

A.Data analyst

B.Data owner

C.Data steward

D.Data custodian

AnswerC

The data steward is accountable within a domain for defining data standards, documenting metadata and monitoring quality. This domain-focused accountability role matches the requirement, unlike executive owners or technical administrators who lack that stewardship mandate.

Why this answer

The data steward is the role primarily accountable for defining data standards and ensuring data quality within a specific domain. This aligns with the DAMA-DMBOK framework, where the data steward acts as the business-side owner of data content, establishing rules for data entry, validation, and metadata management to maintain consistency and accuracy.

Exam trap

The trap here is confusing the data steward with the data owner or data custodian, as many candidates mistakenly think the owner handles domain-level quality or that the custodian defines standards, when in fact the steward is the bridge between business requirements and technical enforcement.

How to eliminate wrong answers

Option A is wrong because a data analyst focuses on querying, analyzing, and reporting data, not on defining standards or governing data quality across a domain. Option B is wrong because a data owner is typically a senior executive accountable for data assets at an enterprise level, not for day-to-day domain-specific standards and quality enforcement. Option D is wrong because a data custodian (or data steward in some frameworks) handles technical implementation, storage, and security, but does not define business-level data standards or quality rules.

Full explanation →

246

MCQeasy

A data analyst creates a dashboard for executives to monitor quarterly sales. Which best practice ensures the dashboard is effective?

A.Place the most important metric in the top-left corner with simple charts.

B.Use a dark background with bright colors for contrast.

C.Include raw data tables for detailed analysis.

D.Use as many charts as possible to show all data.

AnswerA

Executives scan dashboards briefly, so placing the primary sales metric top-left follows natural reading order and guarantees immediate visibility. Simple charts reduce cognitive load, letting decision-makers grasp performance without interpretation effort, satisfying the stem's effectiveness requirement.

Why this answer

Placing the most important metric in the top-left corner leverages the natural reading pattern (left-to-right, top-to-bottom) to immediately draw the executive's attention to the key insight. Using simple charts (e.g., bar or line charts) reduces cognitive load, enabling rapid comprehension of quarterly sales trends without distracting details. This aligns with dashboard design principles that prioritize clarity and actionability over data density.

Exam trap

CompTIA often tests the misconception that more data or flashy visuals improve a dashboard, when in fact effective data communication relies on minimalism and strategic placement of the most critical insight.

How to eliminate wrong answers

Option B is wrong because a dark background with bright colors can cause eye strain and reduce readability, especially in well-lit executive meeting rooms; effective dashboards typically use light backgrounds with high-contrast, accessible color schemes. Option C is wrong because including raw data tables in an executive dashboard defeats its purpose—executives need summarized insights, not granular data, which should be available in a separate drill-down report. Option D is wrong because using as many charts as possible leads to clutter and information overload, obscuring the key sales metrics and making the dashboard ineffective for quick decision-making.

Full explanation →

247

MCQhard

A research firm is acquiring data from public government databases via API. The API rate limits at 100 requests per minute. They need to download 10,000 records, but each request returns a maximum of 100 records. What is the most efficient approach to ensure complete acquisition without being blocked?

A.Use a retry logic with exponential backoff and pagination

B.Request a data dump from the government via email

C.Download one record per second

D.Send all requests simultaneously in parallel

AnswerA

Pagination splits the 10,000 records into 100 requests of 100 records each, satisfying the API's 100-records-per-request cap. Exponential backoff handles the 100-requests-per-minute rate limit by spacing retries after HTTP 429 responses, preventing blocks while ensuring every record is retrieved.

Why this answer

Pagination with retry logic using exponential backoff allows the firm to send requests in a controlled manner, respecting the rate limit and handling potential failures. Sending all requests in parallel would likely exceed the rate limit and cause blocking. Downloading one record per second is too slow.

Requesting a data dump via email is inefficient and may not be supported.

Full explanation →

248

MCQhard

An insurer's data governance team discovers that a legacy reporting pipeline transforms policy effective dates using a hard-coded time zone offset that was correct five years ago but is now wrong for a newly acquired business unit. The team must document where the defect originates, which downstream reports are affected, and how to prevent recurrence. Which governance capability is MOST directly suited to this investigation?

A.Data masking, which obfuscates sensitive fields so non-production environments can be used safely

B.Data lineage, which maps source-to-target transformations and downstream dependencies across the pipeline

C.Data retention scheduling, which automates archival and deletion of records past their required lifecycle

D.Data profiling, which statistically summarizes column distributions to detect outliers and anomalies

AnswerB

Lineage documents how data moves and transforms from source systems through each processing step to consuming reports. That lets the team pinpoint the transformation containing the hard-coded offset and enumerate every downstream report and dashboard that inherits the defect, directly supporting both impact assessment and a targeted fix with regression testing.

Why this answer

Lineage is the capability that records transformation logic and dependency chains, making it the right tool to locate the hard-coded offset and trace every downstream report affected by it. Profiling detects value anomalies without explaining logic, masking protects non-production data, and retention manages lifecycle duration. Only lineage ties the defect's origin to its consumers for impact assessment and regression planning.

Exam trap

The trap here is assuming data profiling can identify transformation logic errors, when profiling only surfaces value-level anomalies.

Full explanation →

249

MCQhard

During a presentation, a stakeholder questions the validity of a correlation found. What is the best response?

A.Correlation does not imply causation, but we can perform further analysis.

B.We can accept the correlation as true.

C.We used a large sample so it's valid.

D.The p-value is low, so it's significant.

AnswerA

Acknowledging that correlation does not imply causation addresses the validity challenge honestly, while offering further analysis keeps the investigation open rather than defensive. This satisfies the stakeholder's concern about the correlation's meaning, distinguishing statistical association from causal mechanism without dismissing the finding outright.

Why this answer

It directly addresses the stakeholder's concern about validity by acknowledging the fundamental statistical principle that correlation does not imply causation. It then proposes a constructive next step—further analysis—which aligns with best practices in data communication, where validating insights requires additional testing (e.g., controlled experiments or causal inference methods). This response demonstrates both technical honesty and a commitment to rigorous data-driven decision-making.

Exam trap

The trap here is that candidates often confuse statistical significance (p-value) or sample size with validity of a correlation, overlooking the core principle that correlation does not imply causation, which is a classic pitfall in data interpretation questions.

How to eliminate wrong answers

Option B is wrong because accepting a correlation as true without scrutiny ignores the possibility of spurious correlations, confounding variables, or sampling bias, which undermines data integrity. Option C is wrong because a large sample size reduces sampling error but does not guarantee that a correlation is meaningful or causal; it can still be due to chance or hidden confounders. Option D is wrong because a low p-value indicates statistical significance (i.e., the correlation is unlikely to be due to random chance), but it does not prove practical importance or causation, and significance can be inflated with large samples.

Full explanation →

250

MCQeasy

In an A/B test, the null hypothesis states that there is no difference between the conversion rates of the control and treatment groups. After collecting data, the p-value is 0.03. Using a significance level α = 0.05, what should the analyst conclude?

A.Reject the null hypothesis; there is a significant difference

B.Accept the alternative hypothesis; the treatment is better

C.The test is inconclusive

D.Fail to reject the null hypothesis; no significant difference

AnswerA

The p-value of 0.03 falls below the 0.05 significance level, so the observed difference is unlikely under the null hypothesis. Rejecting the null is therefore the statistically correct conclusion, indicating a significant difference between control and treatment conversion rates.

Why this answer

Since the p-value (0.03) is less than α (0.05), the null hypothesis is rejected, indicating a statistically significant difference between the groups.

Full explanation →

251

MCQeasy

Which of the following data sources is most likely to generate streaming data?

A.Transactional database

B.Flat file

C.API

D.IoT sensors

AnswerD

IoT sensors emit continuous, time-stamped measurements as events occur, making them a natural streaming source. Unlike batch extracts or static reference tables, sensor telemetry arrives incrementally and is consumed by stream processing platforms in near real time.

Why this answer

Streaming data is continuously generated from sources like IoT sensors, clickstreams, and social media feeds.

Full explanation →

252

MCQmedium

A financial report must be retained for seven years to comply with regulatory requirements. This is an example of which data governance principle?

A.Data freshness

B.Data lineage

C.Data retention

D.Data dictionary

AnswerC

Data retention defines how long records must be kept to satisfy legal or regulatory obligations, directly matching the seven-year requirement. Retention policies specify minimum and maximum storage periods before secure disposal, ensuring the financial report remains available for audits throughout the mandated period.

Why this answer

Data retention policies specify how long data must be kept to meet legal or regulatory obligations.

Full explanation →

253

MCQeasy

A data analyst receives a dataset with a column 'salary' that contains values like '45,000', '55,000', and '65,000'. The analyst notices that the values are stored as text. Which data concept should be applied to convert the salary column from text to numeric format for analysis?

A.Data imputation

B.Data type conversion

C.Data validation

D.Data normalization

AnswerB

Data type conversion casts the text values to a numeric type, stripping the comma separators so arithmetic and aggregation work correctly. The salary column's text storage is the constraint, and conversion changes its type rather than its values.

Why this answer

Data type conversion is the correct concept because the salary values are stored as text (string) but need to be converted to a numeric type (e.g., integer or float) for mathematical operations like aggregation or averaging. In tools like Python (pandas `astype(float)`), SQL (`CAST(salary AS INTEGER)`), or Excel (`VALUE()` function), this explicit conversion ensures the data is treated as numbers, not strings. Without conversion, operations like `SUM` or `AVG` would fail or produce incorrect results.

Exam trap

CompTIA often tests the distinction between data transformation (type conversion) and data preparation techniques like imputation or normalization, trapping candidates who confuse 'changing format' with 'filling gaps' or 'scaling values'.

How to eliminate wrong answers

Option A is wrong because data imputation deals with filling missing values (e.g., using mean or median), not changing the data type of existing values. Option C is wrong because data validation checks whether data meets predefined rules (e.g., range or format constraints), but it does not transform text to numeric format. Option D is wrong because data normalization rescales numeric values to a standard range (e.g., 0–1 or z-scores), which assumes the data is already numeric, not converting text to numbers.

Full explanation →

254

MCQeasy

An analyst is profiling a newly acquired customer table and wants a quick summary of the central tendency and spread of the 'annual_income' column, which contains a few extreme outliers from data entry errors. Which combination of descriptive statistics is most appropriate to report the typical income while limiting the influence of those outliers?

A.Sum and count

B.Median and interquartile range

C.Mean and standard deviation

D.Mode and range

AnswerB

The median is the middle value and is resistant to extreme outliers, so it reflects the typical income even when a few records are corrupted. The interquartile range describes the spread of the middle half of the data and likewise ignores extreme tails. Together they give a robust central tendency and dispersion summary that suits skewed income data with known entry errors.

Why this answer

For skewed data with known outliers, robust statistics are preferred. The median resists extreme values and represents the middle of the distribution, while the interquartile range measures spread using the middle 50 percent of records. Mean and standard deviation, mode and range, and sum and count either amplify outlier influence or fail to describe central tendency and dispersion together.

Exam trap

The trap here is defaulting to mean and standard deviation for any numeric column, even when the scenario states that extreme outliers are present.

Full explanation →

255

MCQmedium

An analyst needs to combine two datasets from different sources that share a common key but have different levels of granularity. Dataset A has daily sales per store, Dataset B has hourly foot traffic per store. The analyst wants to analyze correlation. Which approach is appropriate?

A.Aggregate Dataset B to daily level before merging

B.Use an outer join and keep all rows

C.Disaggregate Dataset A to hourly level by dividing daily sales by hours

D.Join on store and date without aggregation

AnswerA

Correlation requires matching granularity; hourly foot traffic cannot align one-to-one with daily sales rows. Aggregating Dataset B to daily totals creates a shared daily key per store, enabling a valid join and meaningful correlation analysis across the two sources.

Why this answer

Aggregating Dataset B (hourly foot traffic) to the daily level ensures both datasets share the same granularity before merging on the common key (store and date). This allows a valid correlation analysis between daily sales and daily foot traffic without introducing artificial patterns or data duplication. Merging at mismatched granularities would violate the assumption that each row represents a comparable unit of observation.

Exam trap

CompTIA often tests the misconception that disaggregating (splitting) the coarser dataset is acceptable, but this introduces artificial data and violates the assumption of uniform distribution, whereas aggregation preserves the actual measured values.

How to eliminate wrong answers

Option B is wrong because an outer join without aggregation would produce multiple rows per store-date (one for each hour) when joined with daily sales, inflating the number of rows and creating a many-to-one relationship that distorts correlation calculations. Option C is wrong because disaggregating daily sales by simply dividing by hours (e.g., 24) assumes uniform sales distribution, which is rarely true and introduces artificial hourly values that do not reflect actual sales patterns. Option D is wrong because joining on store and date without aggregation retains hourly granularity from Dataset B, causing each daily sales row to repeat for every hour, leading to duplicate data and invalid statistical analysis.

Full explanation →

256

Multi-Selecteasy

A data analyst is designing a dashboard for senior executives who need to quickly monitor key business metrics. Which TWO design principles should the analyst follow? (Choose two.)

Select 2 answers

A.Include detailed data tables for reference

B.Display only the most important KPIs

C.Use consistent formatting and clear labels

D.Add complex interactive filters

E.Use as many colors as possible to make it visually appealing

AnswersB, C

Executives need rapid comprehension, so restricting the dashboard to the most important KPIs reduces cognitive load and directs attention to decision-critical metrics. This satisfies the stem's requirement for quick monitoring by senior executives, since every displayed metric competes for limited attention and dilutes the signal from genuinely vital indicators.

Why this answer

Option B is correct because executive dashboards should surface only the most important KPIs so senior leaders can grasp critical business metrics at a glance without being distracted by secondary data. Option C is correct because consistent formatting and clear labels reduce cognitive load and ensure that executives interpret the displayed KPIs accurately and quickly. Option A is not appropriate because detailed data tables add clutter and slow down high-level decision-making rather than supporting rapid monitoring.

Option D is not appropriate because complex interactive filters increase complexity and require extra effort, which conflicts with the need for quick executive oversight. Option E is not appropriate because using many colors creates visual noise and can obscure meaning instead of improving clarity.

Exam trap

CompTIA often tests the misconception that more data and interactivity always improve a dashboard, when in fact, for executive audiences, simplicity and focus on the most important KPIs are paramount.

Full explanation →

257

MCQmedium

A mid-sized insurance company stores policyholder records in a Microsoft SQL Server database. The compliance team must ensure that any column containing Social Security numbers is masked whenever the data is queried by analysts who are not authorized to view full identifiers. The database administrators want a solution that enforces masking at query time without altering the stored values or requiring application changes. Which SQL Server feature should they implement?

A.Row-Level Security

B.Dynamic Data Masking

C.Always Encrypted

D.Transparent Data Encryption

AnswerB

Dynamic Data Masking applies masking rules at query time based on the user's permissions, so unauthorized analysts see masked values while authorized users see the original data. It does not change stored data or require application changes, which directly satisfies the scenario's requirement to protect Social Security numbers without altering the underlying records.

Why this answer

Dynamic Data Masking is the correct choice because it enforces masking at query time based on user permissions, leaving stored data unchanged and requiring no application modifications. It allows authorized users to see full values while unauthorized users see masked results, which precisely matches the compliance requirement for protecting Social Security numbers in the insurance database.

Exam trap

The trap here is confusing encryption features such as Transparent Data Encryption or Always Encrypted with masking, even though encryption protects data at rest or in use but does not obscure values for authorized queries.

Full explanation →

258

MCQeasy

A data analyst uses Python's pandas library to read a CSV file into a DataFrame. Which function is used to read the file?

A.pd.import_csv()

B.pd.read_excel()

C.pd.load_csv()

D.pd.read_csv()

AnswerD

`pd.read_csv()` parses delimited text into a DataFrame, handling the header row, comma delimiter and type inference automatically. It directly satisfies the stem's requirement to load a CSV file into pandas, unlike `read_excel()` or `read_sql()`, which target different source formats.

Why this answer

pandas reads delimited text files with the top-level function pd.read_csv(), which parses the CSV into a DataFrame and supports parameters like sep, header, dtype, and parse_dates. It is the canonical entry point for CSV ingestion in pandas and is imported as part of the standard `import pandas as pd` idiom. The other options either do not exist in the pandas API or target a different file format.

Exam trap

The trap is that candidates see plausible-sounding function names like import_csv or load_csv and pick them by pattern-matching, rather than recalling that pandas consistently uses the read_* naming convention for I/O.

How to eliminate wrong answers

Option A is wrong because pd.import_csv() is not a pandas function — it is a fabricated name that resembles import statements in other libraries. Option B is wrong because pd.read_excel() reads Excel workbooks (.xls/.xlsx) via openpyxl or xlrd, not CSV files. Option C is wrong because pd.load_csv() does not exist in pandas; 'load' is used in other ecosystems (e.g., numpy.load, json.load) but not for CSV in pandas.

Full explanation →

259

MCQeasy

A data analyst is designing a data model for a sales data warehouse. The model should optimize query performance for aggregations by minimizing joins and duplicating data where necessary. Which schema design should the analyst use?

A.Entity-relationship model

B.Snowflake schema

C.3NF normalized model

D.Star schema

AnswerD

A star schema places a central fact table surrounded by denormalised dimension tables, so aggregation queries join fewer tables and scan pre-joined data. This deliberately duplicates attributes to minimise joins, matching the stated performance requirement.

Why this answer

A star schema is the correct choice because it organizes data into a central fact table surrounded by denormalized dimension tables, which minimizes the number of joins required for aggregation queries. By duplicating dimension attributes rather than normalizing them, the star schema trades storage for query speed, making it ideal for data warehouse workloads that emphasize analytical performance. This design directly supports the requirement to optimize aggregations while reducing join complexity.

Exam trap

DA0-002 often tests the misconception that normalization always improves performance, but in data warehousing, denormalization via star schema is preferred for analytical query speed.

How to eliminate wrong answers

Option A is wrong because an entity-relationship model is a conceptual modeling technique used for OLTP systems, not a physical schema optimized for analytical query performance. Option B is wrong because a snowflake schema normalizes dimension tables into multiple related tables, which increases the number of joins and slows aggregation queries. Option C is wrong because a 3NF normalized model eliminates redundancy and is designed for transactional integrity, not for minimizing joins in analytical queries.

Full explanation →

260

MCQhard

A data analyst is merging two datasets: one containing customer demographics with a 'customer_id' column, and another containing transaction records with a 'cust_id' column. The analyst needs to combine these datasets to analyze purchasing behavior by demographic. Which SQL join condition should be used?

A.ON demographics.customer_id = transactions.cust_id AND demographics.customer_id = transactions.customer_id

B.ON demographics.cust_id = transactions.customer_id

C.ON demographics.customer_id = transactions.cust_id

D.ON demographics.customer_id = transactions.customer_id

AnswerC

This condition correctly matches the 'customer_id' from the demographics table with the 'cust_id' from the transactions table. Despite different column names, they represent the same entity. Using this join condition ensures that each customer's demographic data is linked to their transactions, enabling accurate analysis of purchasing behavior by demographic.

Why this answer

The correct join condition matches the customer identifier from the demographics table with the corresponding identifier in the transactions table, despite the different column names. This allows the analyst to combine the datasets accurately. Using the wrong column names or adding non-existent conditions would result in SQL errors and prevent the merge, so it is essential to map the keys correctly.

Exam trap

The trap here is assuming that column names must match exactly for a join, when in fact they can differ as long as the data values correspond.

Full explanation →

261

MCQmedium

A dashboard designer wants to ensure that the most important metric, such as total revenue, is prominently displayed at the top left. Which dashboard design principle is being applied?

A.Visual hierarchy

B.Consistent color coding

C.Appropriate precision

D.Data-ink ratio

AnswerA

Visual hierarchy arranges elements by importance, so placing total revenue top-left exploits natural reading order to draw attention first. This satisfies the requirement that the most important metric be prominently displayed, guiding the viewer's eye before secondary content.

Why this answer

Visual hierarchy is the design principle of arranging elements so the most important information draws attention first, typically through size, position, color, or contrast. Placing total revenue at the top left leverages the natural reading pattern (F-pattern or Z-pattern) and prominence of that location to signal importance. This is exactly what visual hierarchy governs.

Exam trap

DA0-002 often tests design principles with scenario descriptions; candidates confuse visual hierarchy (importance via placement/size) with data-ink ratio (minimalism) or color coding (consistency), so reading the scenario for 'prominently displayed' is key.

How to eliminate wrong answers

Option B is wrong because consistent color coding is about using the same colors for the same categories across visuals, not about positioning the most important metric. Option C is wrong because appropriate precision concerns how many decimal places or significant digits are shown, not placement. Option D is wrong because data-ink ratio is about minimizing non-data elements like gridlines and decorations, not about where a metric is placed.

Full explanation →

262

MCQhard

A data analyst uses linear regression to model the relationship between advertising spend and sales. The residual plot shows a clear U-shaped pattern. What assumption is violated?

A.Independence of residuals

B.Homoscedasticity

C.Normality of residuals

D.Linearity

AnswerD

A U-shaped residual pattern means the model systematically under- and over-predicts across the predictor range, indicating the true relationship is curved rather than straight. The linearity assumption, that predictors relate to the outcome additively in a straight line, is therefore violated.

Why this answer

The U-shaped pattern in the residual plot indicates that the relationship between advertising spend and sales is not linear; the model fails to capture the curvature in the data. Linear regression assumes a straight-line relationship between predictors and the response, so a systematic pattern like a U-shape directly violates the linearity assumption. This means the model is misspecified and requires a transformation or a nonlinear modeling approach.

Exam trap

CompTIA often tests the distinction between residual pattern shapes and their corresponding assumptions, so the trap here is that candidates confuse a curved pattern (nonlinearity) with heteroscedasticity or non-normality, leading them to pick B or C instead of D.

How to eliminate wrong answers

Option A is wrong because independence of residuals refers to errors being uncorrelated with each other, often violated in time-series data, but a U-shaped pattern does not imply autocorrelation. Option B is wrong because homoscedasticity means constant variance of residuals across fitted values, which would appear as a funnel or cone shape, not a U-shaped curve. Option C is wrong because normality of residuals concerns the distribution of errors (checked via Q-Q plot or histogram), not the pattern of residuals versus fitted values; a U-shaped pattern does not directly indicate non-normality.

Full explanation →

263

MCQhard

A data engineer is profiling a dataset of customer orders and notices that the 'order_date' column contains values in multiple formats: 'YYYY-MM-DD', 'MM/DD/YYYY', and 'DD-Mon-YYYY'. The column is currently stored as a string. Which action should the engineer take to ensure consistent date handling for analysis?

A.Use a CASE statement to convert each format to a standard date type during queries.

B.Replace all delimiters with hyphens to make the strings look uniform.

C.Leave the column as string and rely on the BI tool to interpret the formats.

D.Parse each format and load the dates into a new column with a consistent DATE data type.

AnswerD

Parsing the various string formats and storing the result in a DATE column enforces consistency at the storage layer. This ensures all downstream queries and tools interpret the dates correctly without additional conversion logic. It also enables date-specific functions and comparisons. This is the most reliable way to standardize date data for analysis.

Why this answer

The engineer should parse each date format and load the values into a new column with a consistent DATE data type. This standardizes the data at the storage level, ensuring all analytical queries and tools interpret dates uniformly. It also allows the use of native date functions and avoids repeated conversion logic in every query.

Exam trap

The trap here is thinking that a query-time CASE statement is sufficient, but it leaves the data inconsistent and burdens every future query with conversion logic.

Full explanation →

264

MCQeasy

In a sales database, an analyst needs to retrieve all orders where the order amount is between $100 and $500. Which WHERE clause should be used?

A.WHERE amount IN (100, 500)

B.WHERE amount >= 100 AND amount <= 500

C.WHERE amount BETWEEN 100 AND 500

D.WHERE amount > 100 AND amount < 500

AnswerC

BETWEEN is inclusive on both bounds, returning rows where amount equals 100 or 500 as well as those between. This satisfies the stem's requirement to capture orders from $100 to $500, whereas exclusive comparisons would silently omit boundary orders worth exactly $100 or $500.

Why this answer

The BETWEEN operator is inclusive of the boundary values, so WHERE amount BETWEEN 100 AND 500 will return all orders where the amount is greater than or equal to 100 and less than or equal to 500. This is the most concise and correct way to express the range.

Exam trap

DA0-002 often tests the inclusivity of BETWEEN, and candidates may incorrectly choose strict inequalities or the IN operator. The trap is forgetting that BETWEEN includes the boundary values.

How to eliminate wrong answers

Option A is wrong because IN (100, 500) only matches exactly 100 or 500, not the range in between. Option B is technically correct but not the best answer because it is more verbose; however, in a multiple-choice setting, BETWEEN is the standard and preferred syntax for ranges. Option D is wrong because it uses strict inequalities, excluding orders with amount exactly 100 or 500, which does not meet the requirement of 'between $100 and $500' inclusive.

Full explanation →

265

MCQmedium

A BI analyst is creating a report in Power BI and needs to calculate the total sales for the current year compared to the previous year. Which DAX function should be used to calculate the previous year's sales?

A.SAMEPERIODLASTYEAR

B.PREVIOUSYEAR

C.TOTALYTD

D.DATEADD

AnswerA

SAMEPERIODLASTYEAR is a time intelligence function that shifts the current filter context back exactly one year, returning the prior-year equivalent period. It satisfies the year-over-year comparison requirement by operating on a marked date table, so the previous year's sales total aligns correctly with the current year's.

Why this answer

SAMEPERIODLASTYEAR returns a set of dates in the previous year for the same period, which is ideal for year-over-year comparisons.

Full explanation →

266

MCQmedium

A company requires real-time masking of credit card numbers for customer support agents while allowing full access for accountants. Which technique should be implemented?

A.Dynamic data masking

B.Tokenization

C.Static data masking

D.Data encryption

AnswerA

Dynamic data masking obscures credit card numbers in query results at runtime based on the requesting user's role, so support agents see masked values while accountants retain full access. Static masking would alter stored data permanently, failing the real-time requirement.

Why this answer

Dynamic data masking (DDM) applies masking rules at query runtime based on user privileges, allowing accountants full access while customer support agents see only masked credit card numbers. Unlike static masking, DDM does not alter the underlying stored data, making it ideal for real-time, role-based obfuscation without duplicating or transforming the database.

Exam trap

CompTIA often tests the misconception that encryption or tokenization can provide real-time, role-based masking, but these technologies either require decryption (exposing the full value) or introduce latency and storage overhead, making dynamic data masking the only correct choice for this use case.

How to eliminate wrong answers

Option B (Tokenization) is wrong because it replaces sensitive data with a non-sensitive token stored in a separate vault, requiring a detokenization process that adds latency and is not designed for real-time, role-based masking within the same database. Option C (Static data masking) is wrong because it creates a permanent, masked copy of the data in a non-production environment, which cannot provide real-time, on-the-fly masking for live queries. Option D (Data encryption) is wrong because encryption protects data at rest or in transit but does not provide role-based masking at query time; decryption keys grant full access, not partial masking.

Full explanation →

267

MCQhard

A data architect is designing an ETL pipeline to ingest streaming data from IoT sensors. The data must be available for real-time analytics. Which acquisition method is best?

A.Real-time streaming via API

B.Poll sensors every hour

C.Manually upload sensor logs

D.Batch load daily CSV files

AnswerA

Real-time streaming via API pushes each IoT sensor reading to the pipeline as it is produced, rather than buffering batches. This satisfies the requirement that data be available for real-time analytics, unlike batch acquisition which introduces latency before availability.

Why this answer

Real-time streaming via API is the best method because IoT sensors generate continuous data that must be ingested with sub-second latency for real-time analytics. APIs (e.g., REST, WebSocket, or MQTT) enable event-driven ingestion, allowing the ETL pipeline to process each sensor reading as it arrives, which is essential for time-sensitive use cases like anomaly detection or live monitoring.

Exam trap

The trap here is that candidates may confuse 'real-time' with 'frequent batch' and choose hourly polling (Option B), not realizing that real-time analytics requires sub-second latency, not just periodic updates.

How to eliminate wrong answers

Option B is wrong because polling sensors every hour introduces latency of up to 60 minutes, which violates the real-time analytics requirement and can cause data staleness for time-critical decisions. Option C is wrong because manually uploading sensor logs is not automated, introduces human error, and cannot achieve the low-latency ingestion needed for streaming data. Option D is wrong because batch loading daily CSV files imposes a 24-hour delay, making the data unavailable for real-time analytics and contradicting the explicit requirement for immediate data availability.

Full explanation →

268

MCQhard

A healthcare analytics team is building a predictive model to identify patients at high risk of readmission within 30 days of discharge. The dataset includes 50,000 patient records with 200 features, including demographics, vital signs, lab results, and historical admissions. The target variable is binary (readmitted or not). The team uses a logistic regression model and achieves an AUC of 0.72 on the test set. However, the model's calibration is poor: for patients predicted to have a 70% risk, the actual readmission rate is only 40%. The team wants to improve calibration without significantly reducing discrimination (AUC). The data scientist suggests applying Platt scaling. However, the team lead is concerned that Platt scaling may reduce the model's ability to rank patients correctly. Which of the following is the best course of action?

A.Remove poorly calibrated predictions by discarding all patients with predicted risk between 0.3 and 0.7.

B.Ignore calibration because AUC is the only metric that matters for readmission risk models.

C.Apply Platt scaling on a held-out validation set to recalibrate the predicted probabilities without refitting the original model.

D.Switch to a random forest model, which inherently produces better-calibrated probabilities.

AnswerC

Platt scaling fits a logistic regression on the model's raw scores using a held-out validation set, correcting probability estimates while leaving the underlying model and its ranking untouched. Because it is monotonic, discrimination and AUC are preserved, satisfying the calibration goal without refitting.

Why this answer

Platt scaling is a post-processing technique that fits a logistic regression model on the predicted probabilities from the original model using a held-out validation set. This recalibrates the probabilities without altering the ranking of patients (the AUC remains unchanged), directly addressing the poor calibration while preserving discrimination. Option C correctly describes this procedure.

Exam trap

The trap here is that candidates may think Platt scaling changes the model's ranking (AUC), but in reality it applies a monotonic transformation that preserves rank order, so discrimination is unaffected.

How to eliminate wrong answers

Option A is wrong because discarding patients with predicted risk between 0.3 and 0.7 removes a large portion of the data and does not fix the underlying miscalibration; it merely hides the problem and reduces the model's utility. Option B is wrong because AUC measures only rank ordering, not probability accuracy; for clinical risk models, well-calibrated probabilities are critical for decision-making (e.g., resource allocation). Option D is wrong because random forest models are known to produce poorly calibrated probabilities due to their averaging of decision tree outputs, often requiring their own calibration (e.g., isotonic regression) and do not inherently guarantee better calibration than logistic regression.

Full explanation →

269

MCQeasy

A data analyst is examining a dataset of customer transactions and notices that the 'transaction_amount' column contains negative values. The analyst determines that these negative values represent refunds. Which data quality dimension is most directly relevant to this finding?

A.Consistency

B.Accuracy

C.Validity

D.Completeness

AnswerC

Validity ensures data conforms to defined business rules or constraints. If the business rule expects transaction amounts to be positive (e.g., sales), then negative values violate that rule. However, if refunds are allowed, the rule may need to accommodate negatives. The analyst's realization that negatives represent refunds highlights a validity check: are these values valid according to the expected domain? This makes validity the most relevant dimension.

Why this answer

Validity is about whether data adheres to defined rules or constraints. Negative transaction amounts may be valid if refunds are permitted, but the analyst must verify that such values are allowed by business rules. Accuracy, completeness, and consistency do not directly address the conformity to rules.

Thus, validity is the most relevant dimension when encountering unexpected negative values that might indicate either a data error or a legitimate business process.

Exam trap

The trap here is confusing validity with accuracy. Negative values can be accurate (they correctly represent refunds) but may still violate a validity rule if the system expects only positive amounts.

Full explanation →

270

MCQmedium

A data scientist is performing a hypothesis test with a significance level α=0.05. The p-value obtained is 0.03. What should the scientist conclude?

A.Reject the null hypothesis because the p-value is less than the significance level.

B.Fail to reject the null hypothesis because the p-value is greater than 0.01.

C.The test is inconclusive, need a larger sample size.

D.Accept the null hypothesis because the p-value is small.

AnswerA

With α=0.05, a p-value of 0.03 falls inside the rejection region, so the null hypothesis is rejected. This satisfies the stem's stated significance level and p-value, indicating the observed result is statistically significant at that threshold.

Why this answer

The decision rule for hypothesis testing is: if p-value < α, reject the null hypothesis. Here p = 0.03 and α = 0.05, so 0.03 < 0.05, meaning the result is statistically significant and the null hypothesis should be rejected in favor of the alternative. This indicates the observed effect is unlikely to have occurred by chance alone at the 5% significance level.

Exam trap

DA0-002 often tests the misconception that a small p-value means 'accept the null' or that p-values should be compared to a value other than the stated α — candidates confuse rejection logic or misread the threshold.

How to eliminate wrong answers

Option B is wrong because it compares the p-value to 0.01, which is not the stated significance level — the analyst set α=0.05, and the correct comparison is 0.03 < 0.05, leading to rejection, not failure to reject. Option C is wrong because the test is not inconclusive — the p-value clearly falls below α, so a definitive decision can be made without a larger sample. Option D is wrong because the logic is inverted: a small p-value leads to rejecting the null hypothesis, not accepting it; also, hypothesis tests never 'accept' the null, they only fail to reject it.

Full explanation →

271

MCQmedium

A healthcare analytics team is building a classification model to predict patient readmission within 30 days. The dataset contains 10,000 records with 30 features, including demographics, vital signs, lab results, and medication history. The target variable is imbalanced: 85% no readmission, 15% readmission. The team used logistic regression with default settings and achieved an accuracy of 85%, but the model predicted 'no readmission' for all patients. The lead analyst suspects the model is not learning due to class imbalance. The team has time to implement one corrective action before the next model review. Which action should the team take?

A.Remove features with low variance to reduce noise

B.Apply SMOTE to oversample the readmission class

C.Use accuracy as the evaluation metric to monitor improvement

D.Switch to a random forest model with default settings

AnswerB

SMOTE synthesises minority-class readmission examples, rebalancing the 85/15 split so logistic regression no longer converges on the majority class. This directly counters the imbalance causing the all-negative predictions, addressing the stated constraint within one corrective action.

Why this answer

SMOTE (Synthetic Minority Oversampling Technique) directly addresses the class imbalance by generating synthetic samples for the minority class (readmission). This forces the logistic regression model to learn decision boundaries that separate the two classes, rather than defaulting to the majority class prediction. With 85% majority and 15% minority, accuracy alone is misleading, and SMOTE is a proven technique to improve recall for the minority class.

Exam trap

The trap here is that candidates often choose accuracy as a metric (Option C) because it seems intuitive, but in imbalanced datasets, accuracy is misleading and does not reflect model performance for the minority class.

How to eliminate wrong answers

Option A is wrong because removing low-variance features does not address class imbalance; it only reduces noise or redundant features, but the model will still predict the majority class if the imbalance is not handled. Option C is wrong because using accuracy as the evaluation metric is exactly the problem—it will remain high (85%) even if the model predicts all 'no readmission', so it does not monitor improvement for the minority class. Option D is wrong because switching to a random forest model with default settings does not inherently solve class imbalance; random forest can also be biased toward the majority class without techniques like class weighting or resampling.

Full explanation →

272

Multi-Selectmedium

A data analyst is preparing a dataset for a machine learning model and needs to handle missing values in several columns. The analyst wants to choose appropriate imputation methods. Which TWO of the following are valid considerations when selecting an imputation technique? (Choose two.)

Select 2 answers

A.Imputation should always use the mean for numerical variables to preserve the distribution.

B.The proportion of missing data in a column affects the reliability of imputation.

C.The missing data mechanism (MCAR, MAR, MNAR) influences the choice of imputation method.

D.Imputation should be performed before splitting data into training and test sets to avoid data leakage.

E.The choice of imputation method should be based solely on computational efficiency.

AnswersB, C

A high proportion of missing values (e.g., >50%) can make imputation unreliable and may warrant dropping the column or using advanced techniques. The amount of missingness impacts the confidence in imputed values and the potential for bias. Thus, it is a valid consideration when selecting an imputation method.

Why this answer

The missing data mechanism determines whether imputation can be unbiased, and the proportion of missing data affects reliability. These are key statistical considerations. Mean imputation is not always appropriate, imputation should occur after train-test split to avoid leakage, and efficiency alone is insufficient.

Therefore, the mechanism and proportion are the valid considerations.

Exam trap

The trap here is assuming that mean imputation is always safe and that imputation can be done before splitting; both can lead to biased models.

Full explanation →

273

Multi-Selectmedium

Which TWO of the following are components of time series data?

Select 2 answers

A.Mean

B.Variance

C.Trend

D.Seasonality

E.Median

AnswersC, D

Trend is the long-term directional movement of a series over time, rising, falling or flat, ignoring short-term fluctuations. It is a core time series component alongside seasonality, cyclical variation and irregular residuals, so it satisfies the question's requirement.

Why this answer

Trend (C) is a core component of time series data because it represents the long-term, systematic increase or decrease in the series level over time, which is exactly what decomposition methods (e.g., additive or multiplicative decomposition) isolate. Seasonality (D) is also a core component because it captures regular, calendar-linked repeating patterns (e.g., daily, weekly, monthly, or quarterly cycles) that recur with a fixed period. By contrast, Mean (A), Variance (B), and Median (E) are descriptive statistics or summary measures of a distribution, not structural components of a time series; they describe the data's central tendency or spread rather than the temporal dynamics that decomposition separates.

Exam trap

DA0-002 often tests whether candidates confuse statistical summary measures (mean, median, variance) with the structural components of time series (trend, seasonality, cyclicality, irregular), so candidates must recognize that the question asks for components, not statistics.

Full explanation →

274

Multi-Selecteasy

A data analyst is designing a dashboard for a sales team. Which TWO of the following are best practices for dashboard design?

Select 2 answers

A.Use complex visualizations to impress users.

B.Include as many KPIs as possible on one screen.

C.Use consistent color coding for similar metrics.

D.Place the most important information at the top or left.

E.Use a single chart type for all visuals.

AnswersC, D

Assigning the same colour to equivalent metrics lets the sales team compare values across charts without relearning the legend each time. Consistent colour coding reduces cognitive load and misinterpretation, a core dashboard design principle for recurring operational reporting.

Why this answer

Option C is correct because consistent color coding for similar metrics reduces cognitive load and lets viewers instantly associate a color with a metric's meaning across the dashboard, improving readability and faster interpretation. Option D is correct because placing the most important information at the top or left follows the natural F-pattern/Z-pattern reading flow in left-to-right cultures, ensuring key KPIs are seen first. Option A is wrong because complex visualizations prioritize aesthetics over clarity and can confuse users rather than inform decisions.

Option B is wrong because cramming too many KPIs onto one screen creates clutter and dilutes focus from the metrics that matter most. Option E is wrong because using a single chart type for all visuals ignores data-appropriate visualization choices, such as bars for comparison and lines for trends.

Exam trap

The trap here is that candidates often confuse 'impressive visuals' with effective communication, or assume that more data equals better insights, when in fact simplicity and consistency are the hallmarks of professional dashboard design.

Full explanation →

275

Multi-Selecthard

Which THREE of the following are appropriate ways to handle outliers when communicating data insights?

Select 3 answers

A.Document the outlier and its potential impact in the report.

B.Ignore the outlier and proceed with the analysis.

C.Investigate the cause of the outlier.

D.Use a box plot to visualize the distribution including outliers.

E.Remove the outlier from the dataset to clean the data.

AnswersA, C, D

Documenting an outlier and its potential impact preserves analytical transparency without distorting the underlying distribution, satisfying the requirement to communicate insights honestly. Rather than deleting or capping extreme values, this approach flags them for stakeholders, enabling informed interpretation. It suits reporting scenarios where the outlier may signal genuine business events warranting investigation.

Why this answer

Option A is correct because documenting an outlier and its potential impact preserves transparency and lets stakeholders judge how the anomaly may affect conclusions. Option C is correct because investigating the cause of an outlier determines whether it is a data-entry error, a measurement fault, or a genuine signal that must be retained. Option D is correct because a box plot visualizes the distribution and explicitly displays outliers beyond the whiskers, making them visible rather than hidden.

Option B is not appropriate because silently ignoring an outlier conceals potentially important information and biases the analysis. Option E is not appropriate because automatically removing an outlier without justification can distort the dataset and discard legitimate extreme values.

Exam trap

The trap here is that candidates may think removing outliers is always a standard data cleaning step, but the exam emphasizes that outliers must be investigated and documented rather than automatically deleted, as they can carry significant meaning.

Full explanation →

276

MCQmedium

A retail company stores customer purchase history in a relational database. The database contains a table 'transactions' with columns: transaction_id, customer_id, product_id, quantity, price, and transaction_date. A data analyst needs to create a report that shows total revenue per customer for the last quarter. Which data concept describes the relationship between customer_id and total revenue?

A.Foreign key

B.Composite attribute

C.Derived attribute

D.Atomic attribute

AnswerC

Total revenue is not stored but computed by aggregating quantity multiplied by price grouped by customer_id. Because it is calculated from existing stored columns rather than persisted, it is a derived attribute, satisfying the report's need for per-customer revenue.

Why this answer

Total revenue is calculated by summing (quantity * price) for each customer, making it a derived attribute because it is computed from existing stored data (quantity and price) rather than stored directly. In the context of the 'transactions' table, customer_id is a stored key, but total_revenue is not stored; it is derived via aggregation, which matches the definition of a derived attribute in database design.

Exam trap

CompTIA often tests the confusion between a derived attribute (computed from other attributes) and a foreign key (a referential constraint), leading candidates to incorrectly select 'foreign key' because customer_id appears in multiple tables.

How to eliminate wrong answers

Option A is wrong because a foreign key is a column that references a primary key in another table to enforce referential integrity; customer_id in the transactions table is a foreign key referencing the customers table, but total revenue is not a key—it is a computed value. Option B is wrong because a composite attribute is an attribute that can be divided into smaller sub-parts (e.g., address into street, city, zip); total revenue is a single calculated value, not composed of multiple atomic sub-attributes. Option D is wrong because an atomic attribute is indivisible and stored directly (e.g., price, quantity); total revenue is not stored but derived, so it violates the atomicity principle.

Full explanation →

277

MCQhard

An organization needs to acquire data from a third-party vendor. The data will be used for regulatory reporting. Which of the following should be the primary consideration before acquiring the data?

A.Legal and compliance requirements

B.Volume of data

C.Data format

D.Cost of the data

AnswerA

Legal and compliance requirements determine whether the vendor's data may lawfully be acquired and used for regulatory reporting, covering licensing, privacy and provenance. This satisfies the stem's primary-consideration constraint because non-compliant data invalidates the report regardless of its quality.

Why this answer

When acquiring data for regulatory reporting, legal and compliance requirements must be the primary consideration because the data must adhere to specific laws (e.g., GDPR, HIPAA, SOX) and industry regulations. Failing to ensure compliance can result in legal penalties, fines, or rejection of the report by regulatory bodies. This overrides technical or cost concerns, as non-compliant data is unusable for its intended purpose.

Exam trap

The trap here is that candidates prioritize technical or cost factors (volume, format, price) over the foundational legal and compliance gate, mistakenly assuming any data can be adapted later without verifying regulatory fitness first.

How to eliminate wrong answers

Option B is wrong because the volume of data is a secondary operational concern (e.g., storage, processing bandwidth) but does not address whether the data legally satisfies regulatory mandates. Option C is wrong because data format (e.g., CSV, JSON, XML) is a technical integration detail that can be transformed later, not a primary legal or compliance gate. Option D is wrong because cost is a business negotiation factor; even free data must first meet regulatory requirements to be used for reporting.

Full explanation →

278

MCQmedium

A data analyst wants to understand the relationship between advertising spend and sales revenue. The analyst calculates a Pearson correlation coefficient of 0.85. Which of the following is the best interpretation?

A.There is a strong positive linear relationship between advertising spend and sales.

B.85% of the variation in sales is explained by advertising spend.

C.Increasing advertising spend by $1 will increase sales by $0.85.

D.There is a strong negative linear relationship between advertising spend and sales.

AnswerA

A coefficient of 0.85 sits near the top of the -1 to +1 range, and its positive sign confirms that as advertising spend rises, sales revenue tends to rise too. The magnitude indicates a strong, near-linear association, satisfying the stem's request to interpret the calculated value.

Why this answer

A Pearson correlation coefficient of 0.85 indicates a strong positive linear relationship between the two variables. The value is close to +1, meaning as advertising spend increases, sales revenue tends to increase in a linear fashion. Correlation measures the strength and direction of a linear association, not causation or predictive proportion.

Exam trap

The trap here is confusing the correlation coefficient r with the coefficient of determination r², causing candidates to select the '85% of variation' answer.

How to eliminate wrong answers

Option B is wrong because 85% of variation explained would require squaring the correlation (r² = 0.7225, or ~72%), and even that describes coefficient of determination, not the raw r value. Option C is wrong because correlation does not imply a slope of 0.85; that would require regression coefficients, and correlation is unitless. Option D is wrong because a positive value of 0.85 indicates a positive relationship, not a negative one.

Full explanation →

279

MCQmedium

A data analyst is building a dataset from multiple sources and needs to ensure data quality. During the data acquisition phase, which activity is most important to perform?

A.Data visualization

B.Data cleaning

C.Data profiling

D.Data modeling

AnswerC

Profiling computes null counts, distinct values, ranges and formats across each source column, exposing anomalies before data enters the warehouse. This satisfies the data quality requirement by revealing inconsistencies at acquisition time, when they are cheapest to correct.

Why this answer

Data profiling is the most important activity during the data acquisition phase because it involves examining source data to understand its structure, content, and quality issues before integration. This step identifies missing values, data types, duplicates, and inconsistencies early, preventing downstream errors in analysis. Without profiling, subsequent cleaning and modeling may be based on flawed assumptions about the data.

Exam trap

CompTIA often tests the distinction between data profiling (discovery/assessment) and data cleaning (correction), leading candidates to mistakenly choose cleaning as the first step during acquisition when profiling must come first to identify what needs cleaning.

How to eliminate wrong answers

Option A is wrong because data visualization is a presentation and exploratory analysis technique used after data is acquired and cleaned, not during acquisition. Option B is wrong because data cleaning is a corrective process that typically follows data profiling; performing cleaning without first profiling can waste effort on unknown issues or miss critical quality problems. Option D is wrong because data modeling defines relationships and structures for storage or analysis, which occurs after data is acquired and understood, not during the initial acquisition phase.

Full explanation →

280

MCQhard

A company is analyzing customer feedback sentiment. The dataset is highly imbalanced with 95% positive and 5% negative comments. Which technique should the analyst use to address class imbalance before modeling?

A.Use accuracy as the evaluation metric

B.Undersample the majority class

C.Oversample the majority class

D.Use SMOTE

AnswerD

SMOTE generates synthetic minority-class samples by interpolating between existing nearest neighbours, rebalancing the 95:5 split before training. This satisfies the stem's requirement to address class imbalance, letting the model learn negative-class patterns instead of defaulting to the majority class.

Why this answer

SMOTE (Synthetic Minority Oversampling Technique) is the correct choice because it generates synthetic samples for the minority class (negative comments) by interpolating between existing minority instances, rather than simply duplicating them. This addresses the 95:5 imbalance without the information loss of undersampling or the overfitting risk of naive oversampling.

Exam trap

The trap here is that candidates often confuse oversampling the minority class with oversampling the majority class, or they incorrectly assume that simply using a different evaluation metric (like accuracy) can fix the imbalance problem without modifying the dataset.

How to eliminate wrong answers

Option A is wrong because accuracy is a misleading metric for imbalanced datasets; a model predicting all comments as positive would achieve 95% accuracy but fail to identify any negative comments. Option B is wrong because undersampling the majority class discards a large amount of potentially useful data, which can lead to loss of important patterns and reduced model performance. Option C is wrong because oversampling the majority class would exacerbate the imbalance, making the model even more biased toward the majority class.

Full explanation →

281

Multi-Selecteasy

Which TWO of the following are measures of central tendency?

Select 2 answers

A.Median

B.Range

C.Variance

D.Standard deviation

E.Mean

AnswersA, E

The median is a measure of central tendency because it identifies the middle value of an ordered dataset, dividing it into two equal halves. Unlike the mean, it resists distortion by extreme outliers, satisfying the stem's requirement for a positional average rather than a measure of dispersion or spread.

Why this answer

The median (A) is a measure of central tendency because it identifies the middle value of a dataset when ordered, representing the central point that divides the data into two equal halves. The mean (E) is also a measure of central tendency, calculated as the arithmetic average (sum of all values divided by the number of values), and it indicates the typical or central value of the data. The range (B) is a measure of dispersion, showing the difference between the maximum and minimum values, not a central value.

Variance (C) and standard deviation (D) are both measures of spread that quantify how far data points deviate from the mean, so they do not describe central tendency.

Exam trap

DA0-002 often tests the confusion between measures of central tendency (mean, median, mode) and measures of dispersion (range, variance, standard deviation), so candidates who see 'statistical measure' and pick variance or standard deviation fall into the trap.

Full explanation →

282

MCQeasy

A payroll analyst receives a spreadsheet of employee compensation and needs to share aggregate salary statistics with an external benchmarking vendor. Which action best aligns with data minimization principles?

A.Encrypt the spreadsheet with a shared password and email the password in a separate message

B.Replace employee names with randomly generated identifiers before sending the file

C.Send only the calculated aggregates, such as median and quartile salaries by job family

D.Redact Social Security numbers and bank account fields, then send the remaining compensation rows

AnswerC

Data minimization means disclosing only the personal data necessary for the stated purpose. Since the vendor needs aggregate salary statistics, providing computed aggregates by job family eliminates exposure of individual compensation records entirely. This satisfies the benchmarking need while removing the risk of re-identification or misuse of row-level payroll data, making it the strongest alignment with the principle.

Why this answer

Data minimization requires that personal data processing and disclosure be limited to what is adequate, relevant, and necessary for the purpose. The benchmarking vendor needs statistics, not records, so computing aggregates by job family removes individual exposure while meeting the business need. Techniques that merely obscure or protect row-level data, such as pseudonyms, encryption, or partial redaction, leave unnecessary personal data in scope.

Exam trap

The trap here is equating any privacy-protective technique, such as encryption or pseudonymization, with data minimization even though the underlying row-level data is still disclosed.

Full explanation →

283

MCQmedium

A company's sales report shows revenue of $1.2M, but the calculation method is unclear. What data governance artifact would clarify the definition?

A.Audit trail

B.Data dictionary

C.Row-level security

D.Data lineage

AnswerB

A data dictionary documents each field's precise definition, calculation logic and derivation rules, so the revenue figure's ambiguous method becomes explicit and consistently reproducible. It satisfies the need for a governance artifact that clarifies how a metric is defined.

Why this answer

A data dictionary provides clear definitions, calculations, and sources for metrics.

Full explanation →

284

MCQhard

In Looker Studio, you have a data source with daily sales and a separate data source with marketing spend. You want to create a chart that shows sales and marketing spend on the same axis, but the two sources are not joined natively. Which feature should you use to combine them?

A.Data blending

B.Filters

C.Community connectors

D.Calculated fields

AnswerA

Data blending lets you combine metrics from two separate Looker Studio data sources by joining them on a shared dimension, without a native join. This satisfies the requirement to plot sales and marketing spend on the same axis.

Why this answer

Data blending in Looker Studio lets you combine metrics from two different data sources in a single chart by defining a join key (dimension) shared between them. This is the native mechanism for merging unjoined sources without modifying the underlying data. It is designed exactly for scenarios like combining sales and marketing spend on the same axis.

Exam trap

The trap is assuming calculated fields or filters can combine sources, when only data blending natively joins two unconnected data sources in Looker Studio.

How to eliminate wrong answers

Option B is wrong because filters only restrict rows within a single data source; they cannot merge metrics from two separate sources. Option C is wrong because community connectors are third-party integrations for pulling external data into Looker Studio, not for combining two existing sources within a report. Option D is wrong because calculated fields operate within a single data source and cannot reference fields from another source.

Full explanation →

285

MCQmedium

A data analyst wants to find the top 5 products by total sales amount, but only for products that have been sold more than 50 times. Which SQL query accomplishes this?

A.SELECT product_id, SUM(sales_amount) FROM sales GROUP BY product_id HAVING COUNT(*) > 50 ORDER BY SUM(sales_amount) DESC LIMIT 5

B.SELECT product_id, SUM(sales_amount) FROM sales WHERE COUNT(*) > 50 GROUP BY product_id ORDER BY SUM(sales_amount) DESC LIMIT 5

C.SELECT product_id, SUM(sales_amount) FROM sales GROUP BY product_id HAVING COUNT(*) > 50 ORDER BY SUM(sales_amount) ASC LIMIT 5

D.SELECT product_id, SUM(sales_amount) FROM sales GROUP BY product_id WHERE COUNT(*) > 50 ORDER BY SUM(sales_amount) DESC LIMIT 5

AnswerA

HAVING filters grouped rows after aggregation, so COUNT(*) > 50 removes products with 50 or fewer sales before ranking. ORDER BY SUM(sales_amount) DESC with LIMIT 5 then returns the five highest-revenue products. WHERE cannot reference aggregates, making HAVING the only clause satisfying both the sales-count threshold and the top-5 constraint.

Why this answer

HAVING filters after aggregation, then ORDER BY and LIMIT give the top 5.

Full explanation →

286

MCQmedium

A retail analytics team is acquiring point-of-sale transaction files from 40 different store locations. Each store exports its nightly file with a different column order and slightly different header names (for example, 'TxnID' vs 'Transaction_ID' vs 'TransNo'). The analyst must load all files into a single normalized staging table before transformation. Which data preparation approach best addresses this inconsistency at acquisition time?

A.Create a source-to-target mapping for each store's file layout and apply a schema-on-read extraction that renames and reorders columns into a canonical staging schema.

B.Concatenate all raw files into one text file and parse columns positionally using the order from the largest store.

C.Instruct each store to re-export its file using the corporate header standard before the nightly transfer begins.

D.Load each file as-is into separate tables and let the BI tool join them by matching column names at query time.

AnswerA

A per-source mapping preserves each store's original file while projecting it into one canonical column set, so downstream transformations see a single consistent structure. Because the variation is in column names and ordering rather than values, mapping at read time solves the problem without altering source exports or waiting for stores to standardize. It also documents lineage, which supports later auditing of how each field was sourced.

Why this answer

The core issue is structural inconsistency across many source files, so the fix belongs at the acquisition boundary: define a canonical staging schema and map each source layout into it. Schema-on-read mapping handles renamed and reordered columns without changing upstream systems or delaying ingestion, and it creates documented lineage. Approaches that defer reconciliation to queries or force upstream re-exports either spread complexity or block progress.

Exam trap

The trap here is assuming that column-name differences must be fixed in the source system, when the acquisition layer is responsible for normalizing heterogeneous layouts.

Full explanation →

287

Multi-Selecthard

Which THREE are challenges in acquiring data from external sources? (Select three.)

Select 3 answers

A.Data redundancy

B.Unauthorized access

C.Licensing restrictions

D.Rate limiting

E.Data format inconsistency

AnswersC, D, E

External datasets arrive under vendor or public licences governing permitted use, redistribution and derivative works. These contractual terms constrain whether the data can legally be ingested and stored, making licensing restrictions a genuine acquisition challenge rather than a technical processing concern.

Why this answer

Licensing restrictions (C) are a genuine external-data acquisition challenge because third-party providers impose contractual terms, usage caps, and redistribution limits that can legally block or constrain how the data is ingested and reused. Rate limiting (D) is correct because external APIs and feeds typically enforce request quotas (e.g., HTTP 429 responses, tokens per minute), forcing throttling, retries, or batching that complicate acquisition. Data format inconsistency (E) is correct because external sources deliver data in heterogeneous schemas and encodings (JSON, XML, CSV, proprietary formats, differing date/number conventions), requiring parsing and normalization before use.

Data redundancy (A) is an internal data-quality issue about duplicate records rather than an external acquisition obstacle, and unauthorized access (B) is a security concern affecting data protection, not a challenge inherent to acquiring data from outside sources.

Exam trap

DA0-002 often tests the misconception that generic data-quality issues like redundancy or unauthorized access are external-acquisition challenges, when the exam expects licensing, rate limiting, and format inconsistency.

Full explanation →

288

MCQmedium

A data analyst is creating a dashboard in Looker Studio and needs to combine data from two different data sources using a common field. Which feature should be used?

A.Data blending

B.Community connector

C.Filter

D.Calculated field

AnswerA

Data blending joins two different data sources on a shared dimension key, producing a combined result without prior ETL. Looker Studio's blend feature satisfies the stem's constraint of combining sources via a common field, unlike extracting or joining within a single source.

Why this answer

Data blending in Looker Studio allows you to combine data from multiple data sources into a single chart or table by joining them on a common dimension (key field). This is the correct feature when you need to merge data from two different sources, such as Google Analytics and Google Sheets, using a shared field like date or campaign ID.

Exam trap

The trap here is confusing data blending with calculated fields or filters — candidates often think a calculated field can combine sources, but it only operates within one source; data blending is the only feature that joins multiple sources on a common key.

How to eliminate wrong answers

Option B is wrong because a community connector is used to connect Looker Studio to a custom or unsupported data source via the Community Connectors platform — it does not combine data from multiple sources. Option C is wrong because a filter is used to restrict the data displayed in a chart based on conditions, not to join datasets. Option D is wrong because a calculated field creates a new field based on a formula applied to existing fields within a single data source — it cannot merge data from separate sources.

Full explanation →

289

MCQmedium

A data analyst needs to sample 10% of customers from each of three regions (North, South, Central) to ensure proportional representation. Which sampling method should be used?

A.Systematic sampling

B.Cluster sampling

C.Stratified sampling

D.Simple random sampling

AnswerC

Stratified sampling divides the population into the three region strata, then draws a 10% sample independently within each. This guarantees proportional representation of North, South and Central, satisfying the stem's requirement that each region contributes its correct share rather than leaving representation to chance.

Why this answer

Stratified sampling involves dividing the population into homogeneous subgroups (strata) based on a characteristic (here, region) and then sampling proportionally from each stratum. This ensures that each region is represented in the sample according to its proportion in the population. Simple random sampling does not guarantee proportional representation.

Exam trap

The trap is confusing stratified sampling with cluster sampling; candidates might think that sampling from each region is cluster sampling, but cluster sampling involves selecting entire groups, not sampling within each group.

How to eliminate wrong answers

Option A is wrong because systematic sampling selects every nth element from a list, which does not ensure proportional representation across regions unless the list is ordered in a way that aligns with regions, which is not guaranteed. Option B is wrong because cluster sampling involves dividing the population into clusters and randomly selecting entire clusters, which can lead to underrepresentation of some regions. Option D is wrong because simple random sampling selects individuals randomly from the entire population, which may by chance under- or over-represent certain regions, especially with small sample sizes.

Full explanation →

290

MCQmedium

A data engineer is choosing a storage approach for a new analytics platform. The workload consists of wide, denormalized event tables with dozens of attributes, queries that scan a few columns across billions of rows, and heavy aggregation rather than single-row lookups. Which storage structure is best suited to this workload?

A.A normalized schema with many small tables joined at query time

B.A columnar storage format that stores values of each column contiguously and supports compression

C.A row-oriented relational table with a B-tree index on the primary key

D.A key-value store that maps each event ID to a serialized JSON document

AnswerB

Columnar storage lays each column's values together, so a query that reads only a few attributes touches only those column segments and skips the rest. Contiguous values of the same type also compress far better, reducing I/O for the large aggregations described. This matches the analytical pattern of scanning few columns across many rows.

Why this answer

The described workload reads a small number of columns from very wide tables and performs large aggregations, which is exactly what columnar storage optimizes: reading only the needed column segments and compressing similar values. Row-oriented tables with B-tree indexes favor point lookups, key-value stores favor document retrieval by key, and heavily normalized schemas add join overhead. Columnar layout therefore fits the analytical scan pattern best.

Exam trap

The trap here is equating fast primary-key lookups with fast analytical scans, when indexing a row store does not reduce the columns read during a wide aggregation.

Full explanation →

291

MCQmedium

A retail company's data analyst developed a dashboard for store managers to monitor daily sales performance. The dashboard includes numerous metrics such as sales by hour, product category, employee, and customer demographics, along with trend lines and forecast graphs. Despite the comprehensive data, store managers are ignoring the dashboard because they find it cluttered and confusing. They prefer to rely on their intuition and verbal updates from shift leads. The analyst needs to improve communication of data insights to ensure the dashboard is used effectively. Which of the following actions should the analyst take FIRST?

A.Send the raw data in a spreadsheet instead

B.Simplify the dashboard by focusing on key metrics and using clear visual hierarchy

C.Schedule a training session to explain all metrics

D.Add more data points to provide a comprehensive view

AnswerB

The dashboard fails because excessive metrics create cognitive overload, so managers abandon it. Reducing to key metrics with clear visual hierarchy directly addresses the stated clutter and confusion constraint, making insights scannable and actionable before any deeper redesign or training is attempted.

Why this answer

The core issue is that the dashboard is cluttered and confusing, which directly undermines its usability. Option B addresses this by simplifying the dashboard to focus on key metrics and using a clear visual hierarchy, which is the foundational step in effective data communication. Without first reducing cognitive load, no amount of training or additional data will make the dashboard useful for time-constrained store managers.

Exam trap

The trap here is that candidates may confuse 'comprehensive data' with 'effective communication,' leading them to choose options that add more information (D) or provide raw data (A), rather than recognizing that clarity and focus are the primary drivers of dashboard adoption.

How to eliminate wrong answers

Option A is wrong because sending raw data in a spreadsheet would exacerbate the problem by overwhelming managers with unstructured, granular data, requiring them to perform their own analysis—the opposite of a dashboard's purpose. Option C is wrong because scheduling a training session to explain all metrics assumes the problem is a lack of understanding, not the dashboard's poor design; training a user to navigate a cluttered interface is inefficient and does not fix the root cause. Option D is wrong because adding more data points would increase clutter and confusion, directly contradicting the user feedback that the dashboard is already too complex.

Full explanation →

292

MCQeasy

A data analyst wants to use a Z-score to standardize a dataset. The variable has a mean of 50 and a standard deviation of 10. What is the Z-score for a raw value of 70?

A.0.5

B.20

C.-2

D.2

AnswerD

Applying the Z-score formula (x minus mean, divided by standard deviation) gives (70−50)/10 = 2. This standardised value states the raw score sits two standard deviations above the mean, satisfying the stem's requirement to standardise using the given mean of 50 and standard deviation of 10.

Why this answer

The Z-score formula is Z = (X - μ) / σ, where X is the raw value, μ is the mean, and σ is the standard deviation. Plugging in the given values: Z = (70 - 50) / 10 = 20 / 10 = 2. Thus, the raw value of 70 is 2 standard deviations above the mean, corresponding to a Z-score of 2.

Exam trap

The trap here is confusing the difference between the raw value and the mean (20) with the Z-score, or incorrectly reversing the numerator to get a negative Z-score, which would misrepresent the direction from the mean.

How to eliminate wrong answers

Option A (0.5) is wrong because it results from dividing the standard deviation by the difference (10/20) instead of the correct order, or from misapplying the formula as (μ - X)/σ? Actually (50-70)/10 = -2, not 0.5. Option B (20) is wrong because it is simply the difference between the raw value and the mean (70 - 50 = 20) without dividing by the standard deviation. Option C (-2) is wrong because it reverses the sign, computing (50 - 70)/10 = -2, which would indicate the value is below the mean, but 70 is above the mean of 50.

Full explanation →

293

MCQmedium

A data engineer is designing a database for an online store. The database must enforce strict consistency and handle complex transactions involving multiple tables, such as orders, customers, and inventory. Which database type should the engineer choose?

A.Relational database

B.Graph database

C.Key-value store

D.Document database

AnswerA

Relational databases use ACID transactions, ensuring strict consistency and supporting complex joins across normalized tables. They are ideal for transactional systems like order processing where data integrity is critical. This makes relational the best fit for the scenario.

Why this answer

The correct answer is Relational database because it provides ACID compliance and supports complex joins, which are essential for handling transactions across multiple related tables. Other database types lack the necessary transactional guarantees or relational capabilities for this use case.

Exam trap

The trap here is assuming that any database that can store data is suitable; however, only relational databases provide the strict consistency and multi-table transaction support required for this scenario.

Full explanation →

294

MCQeasy

Which chart type is best for visualizing the correlation between two continuous variables?

A.Pie chart

B.Bar chart

C.Scatter plot

D.Line chart

AnswerC

A scatter plot places one continuous variable on each axis, so each point represents a paired observation and the resulting cloud reveals the direction, strength and form of correlation. Line charts suit trends over time, and bar charts suit categorical comparisons.

Why this answer

A scatter plot places each observation as a point on a two-dimensional plane defined by the two continuous variables, making the strength, direction, and shape (linear vs. nonlinear) of their relationship directly visible. It is the standard chart for bivariate correlation analysis.

Exam trap

The trap here is confusing 'correlation between two continuous variables' with 'trend over time' — candidates pick line chart because it also uses two axes, but line charts require an ordered sequence.

How to eliminate wrong answers

Option A is wrong because a pie chart shows parts of a whole for a single categorical variable and cannot represent two continuous variables or their relationship. Option B is wrong because a bar chart compares a categorical dimension against a numeric measure, not two continuous variables against each other. Option D is wrong because a line chart is designed for trends over an ordered sequence (typically time), not for showing the joint distribution or correlation between two continuous variables.

Full explanation →

295

Multi-Selecthard

A data engineer is acquiring semi-structured JSON event logs from a mobile application. Each event contains nested objects and arrays, and some fields are missing depending on event type. Before loading into a relational warehouse, the engineer must flatten and validate the data. Which TWO acquisition practices are most appropriate for this semi-structured source? (Choose two.)

Select 2 answers

A.Validate required fields, data types, and allowed enumerations during ingestion, and quarantine records that fail validation for later review.

B.Store each raw JSON document as a single text column and defer all parsing to report-level SQL using string functions.

C.Allow the ingestion job to infer types from each batch and create new columns automatically whenever an unfamiliar key appears.

D.Load only the fields that are present in every event type and discard optional fields to guarantee a uniform structure.

E.Define an explicit target schema and use schema-on-read extraction that maps nested JSON paths to flat columns, applying defaults or nulls for missing fields.

AnswersA, E

Because JSON allows missing keys and mixed types, ingestion-time validation catches malformed records before they pollute the warehouse. Quarantining failures preserves evidence for debugging and reprocessing once the upstream bug is fixed. Enforcing enumerations on event names prevents silent schema drift, and separating valid from invalid records keeps downstream transformations and aggregates trustworthy.

Why this answer

Semi-structured acquisition works best when the pipeline imposes a governed target schema and validates records at ingestion. Mapping nested paths to flat, typed columns with explicit null handling makes missing fields intentional, while validation with quarantine protects the warehouse from malformed data and schema drift. Deferring parsing to reports, discarding optional fields, or auto-inferring schemas all trade short-term convenience for long-term inconsistency and lost information.

Exam trap

The trap here is treating JSON flexibility as a reason to postpone structure, when the acquisition layer should still enforce a target schema and validate records.

Full explanation →

296

MCQeasy

A data analyst receives a file with the extension .json. This file contains product information with attributes that vary between records. How should this file be classified?

A.Semi-structured data

B.Structured data

C.Transactional data

D.Unstructured data

AnswerA

JSON stores data in key-value pairs with a defined syntax but no rigid schema, so attributes can vary between records. That self-describing yet non-tabular structure is the defining characteristic of semi-structured data, distinguishing it from structured relational tables and unstructured raw content.

Why this answer

A JSON file with varying attributes per record is a classic example of semi-structured data. Unlike strictly structured data (e.g., a relational table with fixed columns), JSON allows each object to have a different set of key-value pairs, making it schema-flexible. This self-describing nature, where metadata is embedded within the data itself, is the defining characteristic of semi-structured formats.

Exam trap

The trap here is that candidates confuse 'structured data' with any data that has a format or organization, forgetting that structured data specifically requires a fixed, predefined schema enforced at write time, unlike JSON's flexible schema-on-read approach.

How to eliminate wrong answers

Option B is wrong because structured data requires a rigid, predefined schema (like a SQL table with fixed columns and data types), which JSON explicitly does not enforce. Option C is wrong because transactional data refers to records of business events (e.g., sales, orders) and is a classification by use case, not by format; a JSON file can contain transactional data, but the question asks how the file itself should be classified based on its structure. Option D is wrong because unstructured data lacks any internal structure or metadata (e.g., raw text, images, audio), whereas JSON has a clear hierarchical structure with keys and values.

Full explanation →

297

MCQeasy

A data analyst notices that the sales numbers in a report differ from the numbers in the finance department's spreadsheet. This discrepancy is most likely due to a lack of:

A.Row-level security

B.Data lineage

C.Data dictionary

D.Single version of truth

AnswerD

Differing sales figures arise when each department maintains its own copy of data, so definitions and refresh timings diverge. A single version of truth establishes one governed, authoritative source that all reports and spreadsheets reference, eliminating the reconciliation discrepancy.

Why this answer

A single version of truth means all departments use the same centralized data, avoiding inconsistencies.

Full explanation →

298

MCQmedium

A data analyst notices that a dataset of customer ages has several missing values. Which method for handling missing data is most appropriate if the data is missing completely at random and the analyst wants to preserve sample size?

A.Forward-fill using the previous value

B.Impute with the mean age

C.Replace missing values with zero

D.Delete all rows with missing data

AnswerB

Mean imputation replaces each missing age with the variable's average, retaining every record and therefore preserving sample size. Because the data is missing completely at random, the missingness is unrelated to any variable, so mean substitution introduces minimal bias compared with deletion methods.

Why this answer

When data is missing completely at random (MCAR) and the analyst wants to preserve sample size, mean imputation is the standard approach — it replaces missing values with the average of the observed values, retaining all rows and avoiding the bias that deletion would introduce. For MCAR data, mean imputation produces unbiased estimates of the mean (though it reduces variance).

Exam trap

DA0-002 often tests the trade-off between preserving sample size and introducing bias — candidates pick deletion for 'cleanliness' or zero-fill for simplicity without considering the distortion each introduces.

How to eliminate wrong answers

Option A is wrong because forward-fill is appropriate for time-series or ordered data where the previous value is a reasonable proxy — customer ages have no inherent order, so forward-fill would introduce arbitrary values. Option C is wrong because replacing missing ages with zero is nonsensical (age zero is a newborn) and would severely distort the distribution and any downstream statistics. Option D is wrong because deleting rows with missing data reduces sample size and, if the missingness is not truly random, introduces selection bias — the question explicitly states the analyst wants to preserve sample size.

Full explanation →

299

MCQmedium

A dataset contains a 'salary' column. The analyst wants to identify outliers using the IQR method. If Q1 = 40,000 and Q3 = 70,000, what is the upper threshold for a non-outlier?

A.130,000

B.85,000

C.115,000

D.100,000

AnswerC

The IQR is 30,000 (70,000 − 40,000), and the upper fence is Q3 + 1.5 × IQR = 70,000 + 45,000 = 115,000. This satisfies the stem's request for the upper non-outlier threshold, so any salary above 115,000 is flagged as an outlier.

Why this answer

The IQR method defines the upper fence as Q3 + 1.5 x IQR. Here IQR = 70,000 - 40,000 = 30,000, so 1.5 x IQR = 45,000 and the upper threshold is 70,000 + 45,000 = 115,000. Any salary above 115,000 is considered an outlier.

Exam trap

The trap is using the wrong multiplier or adding IQR to Q1 instead of Q3; candidates frequently compute Q3 + IQR or Q3 + 2 x IQR and land on 100,000 or 130,000.

How to eliminate wrong answers

Option A is wrong because 130,000 would result from adding 2 x IQR (60,000) to Q3, which is not the standard 1.5 multiplier. Option B is wrong because 85,000 is only 15,000 above Q3, which corresponds to 0.5 x IQR, far too strict and not the conventional threshold. Option D is wrong because 100,000 equals Q3 + IQR (30,000), which is the 1.0 multiplier, not the standard 1.5 used in Tukey's fences.

Full explanation →

300

MCQmedium

An organization is implementing a data warehouse to support business intelligence reporting. The data warehouse must ensure that transactions are processed reliably. Which property guarantees that each transaction is treated as a single, indivisible unit?

A.Consistency

B.Isolation

C.Atomicity

D.Durability

AnswerC

Atomicity guarantees a transaction executes entirely or not at all, treating it as one indivisible unit. If any statement fails, the whole transaction rolls back, leaving no partial writes. Consistency, isolation and durability address different guarantees, so atomicity satisfies the stem's requirement.

Why this answer

Atomicity (option C) is the correct property because it ensures that a transaction is treated as a single, indivisible unit of work. In the context of a data warehouse, this means that either all operations within the transaction are committed successfully, or none are applied, preventing partial updates that could corrupt the data. This is a core component of the ACID (Atomicity, Consistency, Isolation, Durability) model, which is fundamental to reliable transaction processing in databases like SQL Server, Oracle, or PostgreSQL.

Exam trap

The trap here is that candidates often confuse atomicity with consistency, thinking that 'indivisible unit' means the data must be consistent, but consistency is a separate property that ensures data integrity rules are met, not that the transaction is all-or-nothing.

How to eliminate wrong answers

Option A (Consistency) is wrong because consistency ensures that a transaction brings the database from one valid state to another, preserving all defined rules (e.g., constraints, triggers), but it does not guarantee that the transaction is treated as a single unit. Option B (Isolation) is wrong because isolation controls how transaction changes are visible to other concurrent transactions, preventing dirty reads and other anomalies, but it does not address the indivisibility of the transaction itself. Option D (Durability) is wrong because durability guarantees that once a transaction is committed, its changes persist even in the event of a system failure (e.g., via write-ahead logging), but it does not ensure the transaction is atomic.

Full explanation →

Page 4 of 14

All pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Practice DA0-002 by domain

Target a specific domain to shore up weak areas.

Data Acquisition and Preparation Data Concepts and Environments Data Analysis Visualization and Reporting Data Governance

See all domains with question counts →