RoleMath Study Track for CompTIA Data+ (DA0-002) (DA0-002)
A free study companion keyed to the officially published exam domains of CompTIA Data+ (DA0-002) (DA0-002): what each domain covers in plain language, clearly labeled free resources, a guided lab outline for every domain, and interactive self-checks from our own question bank. CompTIA Data+ (DA0-002) exam objectives and certification page
A free, source-cited study companion built on CompTIA's published Data+ (DA0-002) exam objectives, for independent study only. It is not official training, is not affiliated with or endorsed by CompTIA, and is not a pass guarantee. Data+ is an entry-to-early-career data analytics certification - CompTIA recommends roughly 18-24 months of hands-on data or business-analytics experience - not a data-science, data-engineering, or programming credential; no programming is required, though spreadsheet literacy is assumed and SQL and statistics fluency shorten preparation. Every hands-on lab uses free tools and only PUBLIC or SYNTHETIC data with no cloud cost, and you must never upload real PII, PHI, financial, regulated, or employer data to any third-party tool. Verify the current objectives on the official page before your exam.
A complete free CompTIA Data+ (DA0-002) program pinned to the currently published exam objectives, sequenced the way an analytics project actually runs - understand the data, acquire and clean it, analyze it, visualize and report it, then govern it - with every hands-on lab built on free tools and only PUBLIC or SYNTHETIC data at no cloud cost, an end-to-end capstone that carries one public dataset through all five stages, and an explicit recheck of the official objectives before any exam scheduling. This is an entry-to-early-career analytics certification, not a data-science or data-engineering credential: no programming is required, spreadsheet literacy is assumed, and SQL and basic statistics fluency shorten preparation. True beginners can add optional foundational prep before Domain 2.
This draft exposes RoleMath’s authored sequence and evidence plan. The current labs are guided outlines, not yet a fully fixture-backed course, and objective-leaf coverage has not passed the gold-standard gate. Completion does not predict an exam result.
Modules
5
Labs
5
Concept checks
10
Resource mix
2 official / 5 community
Choose an outcome
Three routes through the same evidence
Choose provisionally. Change routes when the work tells you something new about fit, time, or readiness.
Certification-focused
Learners who already have some data or business-analytics exposure and want one current, dependency-ordered DA0-002 sequence across all five domains, with every lab run on free tools and public or synthetic data and a recheck of the official objectives before scheduling.
Completion emphasis: Complete every module, run each lab on free tools with public or synthetic data only (never real regulated data), correct every missed check, finish the end-to-end analytics capstone on a single public dataset, and diff the current objectives before booking the exam - never inferring a score from coverage.
Required phases: Data concepts, types, structures, and environments, Acquisition, preparation, and analysis - the analytical core, Visualization, reporting, governance, and the end-to-end capstone
Analyst skills first
Career changers who want reviewable evidence that they can classify data, clean and join it, compute honest statistics, build an audience-fit dashboard, and handle sensitive fields responsibly, whether or not they sit the exam soon.
Completion emphasis: Retain a labeled artifact per domain - a data dictionary, a cleaned-and-joined dataset with a quality log, a descriptive-statistics analysis, an audience-fit dashboard, and a governance packet on synthetic data - plus the end-to-end capstone, all produced with free tools and public or synthetic data.
Required phases: Data concepts, types, structures, and environments, Acquisition, preparation, and analysis - the analytical core, Visualization, reporting, governance, and the end-to-end capstone
Career-fit sprint
Learners deciding whether data analytics - classifying, cleaning, analyzing, and visualizing data to answer questions - is a direction worth deeper investment before committing to the full DA0-002 exam preparation.
Completion emphasis: Complete the diagnostic, the concepts-and-environment foundation, and the acquisition-and-analysis phase with their labs on public data, then choose a next analytics experiment or a full exam commitment rather than inferring job readiness or a pass from partial coverage.
Required phases: Data concepts, types, structures, and environments, Acquisition, preparation, and analysis - the analytical core
Start safely
Prerequisite diagnostic
Confirm you have the spreadsheet and data comfort Data+ assumes and can install free local tools before the labs; this diagnostic is not a CompTIA prerequisite, a cost promise, or an exam prediction, and DA0-002 is an entry-to-early-career analytics exam that assumes roughly 18-24 months of data or business-analytics exposure.
Are you comfortable working in a spreadsheet - sorting, filtering, writing basic formulas like AVERAGE and SUM, and reading a table of data - since Data+ assumes this baseline and does not re-teach it?
Ready when: Yes, or you will spend a little time on spreadsheet fundamentals before Domain 2 rather than learning the baseline and the analytics skills at once.
If not yet: Spend time on free spreadsheet fundamentals (Khan Academy or a free Sheets/Excel tutorial) first, then start the concepts foundation.
Do you have some hands-on exposure to working with data or business analytics (CompTIA recommends roughly 18-24 months), or are you genuinely new to it?
Ready when: Yes, you have handled real datasets before, or you will take the optional foundations-prep phase (SQLBolt, Kaggle Learn, Khan Academy) before Domain 2.
If not yet: Take the optional foundations-prep phase to build SQL and statistics fundamentals before the acquisition and analysis domains, since those assume this exposure.
Are you comfortable, or willing to become comfortable, with basic SQL (SELECT, GROUP BY, JOIN) and basic statistics (mean, median, standard deviation), since they shorten preparation even though no programming is required?
Ready when: Yes, or you will build SQL fluency with SQLBolt and Kaggle Learn and statistics with Khan Academy in parallel with the domains.
If not yet: Work through SQLBolt and Khan Academy Statistics in the foundations-prep phase before the analysis domain; no programming is required, but SQL and stats fluency make the labs far easier.
Can you install free local tools on your own machine - DB Browser for SQLite and a free spreadsheet (and optionally Power BI Desktop) - or will you use a free hosted tool with public data only?
Ready when: Yes, with permission to install free desktop apps, or a plan to use Google Looker Studio and a free online spreadsheet with public data only.
If not yet: Use the free hosted routes (Looker Studio, an online spreadsheet) with public data only until you can install local tools; every lab has a hosted or local path.
Do you understand that every lab uses only PUBLIC or SYNTHETIC data, and that you must never upload real PII, PHI, financial, regulated, or employer data to any third-party tool such as Looker Studio or an online SQL sandbox?
Ready when: Yes, and you will keep sensitive data off hosted tools, use the synthetic dataset for the governance lab, and prefer local tools whenever data could be sensitive.
If not yet: Review the data-safety baseline and the governance domain first, and do not connect any real data to a hosted tool during these labs.
Have you chosen a pace whose weekly hours you can realistically protect across roughly 40 to 150 total hours depending on your prior data, SQL, and statistics experience?
Ready when: Yes, with a pace selected and the objective recheck, the labs, and the capstone left uncompressed.
If not yet: Pick the steady pace and add the foundations-prep phase if you are newer to data work; reserve the intensive pace for learners with real analytics, SQL, and statistics experience.
Plan, then adapt
Pace options
Steady
16 weeks 7-9 hours/week
A planning estimate of roughly 120-150 hours for a learner who is newer to data work: the optional foundations-prep phase plus one domain block at a time, every lab run on free tools with public or synthetic data, the end-to-end capstone, and an objectives recheck before scheduling.
Standard
12 weeks 7-9 hours/week
A planning estimate of roughly 80-100 hours for a learner with an IT background but limited SQL: the CompTIA-cited domain study paired with one retained artifact per domain, the end-to-end capstone, the missed-check corrections, and an objectives-diff block before any exam logistics.
Intensive
8 weeks 6-8 hours/week
Roughly 40-60 hours for an experienced learner with real data or business-analytics exposure plus SQL and statistics fluency; do not compress the data-quality logging, the descriptive-versus-inferential judgment, the honest-design review, the synthetic-only governance rule, or the end-to-end capstone.
Evidence-gated sequence
Program roadmap
1
Optional foundations: SQL, statistics, and spreadsheet fluency for true beginners
For learners genuinely new to data work, build the assumed baseline before the graded domains: basic SQL (SELECT, GROUP BY, JOIN) with SQLBolt and Kaggle Learn, basic statistics (mean, median, standard deviation, distributions) with Khan Academy, and spreadsheet fluency. This phase is optional and is not part of the DA0-002 objectives; skip it if you already have data exposure.
Exit evidence
Work through enough SQLBolt to write SELECT, WHERE, GROUP BY, and JOIN queries without a reference, since Domains 2 and 3 assume this.
Work through Khan Academy Statistics until mean, median, standard deviation, and the descriptive-versus-inferential distinction are comfortable.
Confirm spreadsheet fluency (sorting, filtering, AVERAGE/MEDIAN/STDEV) so the analysis cross-checks are straightforward.
This phase carries no authored checks; it exists so true beginners meet the baseline the graded domains assume before the acquisition and analysis labs.
2
Data concepts, types, structures, and environments
Build the vocabulary and mental model everything else assumes (Domain 1): classify data types (categorical, discrete, continuous), tell structured from semi-structured and unstructured data, know where data lives (databases, warehouses, lakes, files), and recognize the common tool landscape - then prove it by classifying a real public dataset by hand and producing a data dictionary.
Complete the Domain 1 lab: load the public UCI Iris dataset into local SQLite, inspect storage types with typeof(), classify each column's analytical data type, and produce a completed data dictionary that separates analytical type from storage type.
Complete the spreadsheet-literacy, data-exposure, SQL/stats, free-tools, and data-safety diagnostics, choose a pace you can protect, and be able to explain why a numeric-looking ID is categorical and must not be averaged.
Retain the data dictionary artifact, built on public data with free local tools, with no real or sensitive data used.
Attempt every authored Data concepts and environments check and correct each miss against its cited source before moving to acquisition.
3
Acquisition, preparation, and analysis - the analytical core
Do the work that carries the most exam weight and the most real effort: acquire and clean public data, handle duplicates and missing values, join tables correctly, and derive an analysis-ready dataset (Domain 2), then compute descriptive statistics and SQL aggregates, flag outliers, and distinguish descriptive from inferential reasoning (Domain 3) - all on public data with free tools.
Complete the acquisition-and-preparation lab: import two years of the public World Happiness Report into SQLite, find duplicates and missing values, left-join to expose dropped countries, derive a year-over-year column, export an analysis-ready CSV, and log every data-quality issue.
Complete the analysis lab: compute count/min/max/range/mean and per-group aggregates on public Gapminder-style data, flag outliers beyond two standard deviations, cross-check the mean/median/standard deviation in a spreadsheet, and classify each finding as descriptive or inferential.
Retain a cleaned-and-joined dataset with a quality log and a descriptive-statistics analysis, both produced on public data with free tools and no sensitive data.
Attempt every authored Data acquisition/preparation and Data analysis check and correct each miss against its cited source, tracing why each join, statistic, or outlier decision matters.
4
Visualization, reporting, governance, and the end-to-end capstone
Turn analysis into communication and then handle data responsibly: build an audience-fit dashboard on public data with the right chart types and an honest-design review (Domain 4), then classify PII/PHI, write a lineage note and a data-quality health check, and draft an anonymization plan on a SYNTHETIC dataset (Domain 5) - then integrate all five domains in the end-to-end capstone on a single public dataset.
Complete the visualization lab: build a dashboard on PUBLIC data in Looker Studio (or Power BI Desktop) with a labeled time-series, a sorted zero-based bar chart, a scorecard, and a filter, plus a plain-English executive summary, and pass the honest-design review.
Complete the governance lab on the SYNTHETIC patient-appointment dataset only: classify each field public/PII/PHI, write a lineage note and a multi-dimension quality check, and draft a three-bullet anonymization plan - with no real regulated data used anywhere.
Complete the end-to-end analytics capstone that carries one public dataset through concepts, acquisition, analysis, visualization, and governance, with a governance step on a synthetic overlay, and retain a labeled artifact per domain.
Attempt every authored Visualization-and-reporting and Data-governance check, correct each miss against its cited source, and diff the current CompTIA objectives before choosing a continue, practice, defer, analytics-experiment, or exam-scheduling next decision.
Before a lab
Environment, access, and safety
Required and optional setup
Required
A browser plus text and spreadsheet tools for the CompTIA-cited objectives and for recording each lab's data dictionaries, quality logs, analyses, dashboards, and governance packets
Free local tools: DB Browser for SQLite for the database labs and a free spreadsheet (LibreOffice Calc offline or Google Sheets with public data only) for the statistics cross-checks
A free business-intelligence tool for the visualization lab: Google Looker Studio (hosted, free account, public data only) or Power BI Desktop (free, local, Windows) for a fully offline route
Only PUBLIC or SYNTHETIC data: public open datasets (UCI, World Happiness, Gapminder) for the concepts/acquisition/analysis/visualization labs and the RoleMath synthetic patient-appointment dataset for the governance lab
Optional
Optional foundational prep on free platforms for true beginners - SQLBolt for SQL, Kaggle Learn for cleaning/analysis/visualization, and Khan Academy for statistics - taken before the graded domains
DB Browser for SQLite to inspect the synthetic governance dataset as a table if you prefer SQL to a spreadsheet
Power BI Desktop as a fully local alternative to hosted Looker Studio whenever a dataset could be sensitive
Accounts and accessibility routes
Accounts
The core local route requires no account and no payment: DB Browser for SQLite, LibreOffice Calc, and Power BI Desktop are free and run offline, and every dataset is public or synthetic.
Some hosted resources may require a free account: Google Looker Studio needs a free Google account, and Kaggle Learn needs a free Kaggle account (verify the free label before relying on it).
No lab requires a paid subscription, a cloud account, or a card; if a hosted tool ever moves a needed step behind a paywall, use the local route (DB Browser, LibreOffice, Power BI Desktop) instead.
Equivalent routes
Every database lab has a spreadsheet equivalent (LibreOffice Calc offline) and every hosted step (Looker Studio) has a local equivalent (Power BI Desktop), so account, device, or bandwidth constraints never block a lab; the governance lab is plain-text and fully offline.
All work is keyboard-operable: SQL is typed in DB Browser, spreadsheet formulas are keyboard-entered, and the data dictionaries, checklists, and governance packets are plain Markdown with labeled headings and tables a screen reader can navigate.
In low-bandwidth conditions use the offline routes (DB Browser, LibreOffice Calc, Power BI Desktop), download each public dataset once, and record every artifact in a local document; when a hands-on step cannot run at all, study the objective and record a written expected-state walkthrough labeled simulated.
Safety baseline
Use ONLY public or synthetic data in every lab: public open datasets (UCI, World Happiness, Gapminder) for concepts, acquisition, analysis, and visualization, and the synthetic fake-patient dataset for governance.
Never upload real PII, PHI, financial, regulated, customer, or employer data to any third-party or hosted tool - Looker Studio, an online SQL sandbox, an online spreadsheet, or any hosted BI service; use the local tools for anything that could be sensitive.
The governance lab's data MUST be entirely synthetic - fabricated names, fake emails, invented record numbers; never practice data classification or anonymization on real regulated records.
Prefer local tools (DB Browser for SQLite, LibreOffice Calc, Power BI Desktop) whenever in doubt about data sensitivity, and confirm a dataset is public before connecting it to any hosted tool.
These labs are learning exercises on fabricated or public data; the governance packet is not legal or compliance advice and must never be repurposed as a template filled with real regulated data.
Show your work
Module evidence and missed-check protocol
Module exit evidence
A labeled artifact per domain tied to its module: a classified data dictionary (Domain 1); a cleaned-and-joined dataset with a data-quality log (Domain 2); a descriptive-statistics analysis with a spreadsheet cross-check and outlier note (Domain 3); an audience-fit dashboard with a design review (Domain 4); or a governance packet with PII/PHI classification, a lineage note, a quality check, and an anonymization plan on synthetic data (Domain 5).
A plain-language explanation of the analytical choice, the data type or statistic or chart or classification involved, why the alternative was weaker, and - for the visualization and governance work - which data-safety rule kept the exercise on public or synthetic data.
All authored checks for the domain attempted, with each miss corrected against its cited source and re-applied to a fresh scenario, plus a note confirming every lab used only public or synthetic data with free tools.
After a missed check
Identify whether the question tests data concepts and environments, data acquisition and preparation, data analysis, visualization and reporting, or data governance before reviewing the answer.
Write why the distractor was plausible and which data-type rule, cleaning or join judgment, descriptive-versus-inferential distinction, chart-choice or honest-design guardrail, or PII/PHI classification distinguishes the correct answer.
Change one detail - the data type, the join key, the distribution, the audience, or the sensitivity level - and explain whether the correct answer changes.
Completing this policy demonstrates current-objectives Data+ coverage and hands-on analytics practice inside RoleMath on public and synthetic data; it does not predict an exam score, establish professional data-analytics experience, confer any authorization to handle real regulated data, or serve as a RoleMath credential.
Integrated practice
End-to-end analytics project on a single public dataset, with a synthetic governance overlay
Carry one public dataset through the whole analytics lifecycle - understand it, acquire and clean it, analyze it, visualize and report it, and govern it - integrating all five DA0-002 domains into one reviewable evidence packet, using free tools and only public or synthetic data.
Workflow
Choose one PUBLIC dataset (for example a World Happiness, Gapminder, or open-government CSV) and a question it can answer, then build a data dictionary classifying every column's analytical data type, storage type, and structure (Domain 1).
Acquire and prepare the data in SQLite with DB Browser: detect and handle duplicates and missing values, join any related tables correctly (watching join grain), derive at least one analysis-ready column, export the cleaned dataset, and log every data-quality issue you found and fixed or flagged (Domain 2).
Analyze the prepared data: compute descriptive statistics (central tendency and spread) and grouped SQL aggregates, flag outliers beyond two standard deviations, cross-check the mean/median/standard deviation in a spreadsheet, and classify each finding as descriptive or inferential without over-claiming (Domain 3).
Visualize and report: build an audience-fit dashboard on the public data in Looker Studio (or Power BI Desktop) with fit-for-data chart types, an interactive control, and a plain-English executive summary, then pass the honest-design review (Domain 4).
Govern: build a synthetic overlay for any sensitive fields the dataset would have (or use the provided synthetic patient dataset), classify each field public/PII/PHI, write a data-lineage note tracing the pipeline you just built, run a data-quality health check, and draft a three-bullet anonymization plan (Domain 5) - all on synthetic data.
Assemble the evidence packet: the data dictionary, the cleaned dataset and quality log, the analysis with cross-check, the dashboard and executive summary, and the governance packet, each labeled with the free tool used and confirmed to use only public or synthetic data.
Review the whole project for honesty: confirm no chart is misleading, no descriptive summary is presented as a population conclusion, no correlation is stated as causation, and no real regulated data or sensitive upload appears anywhere in the packet.
Diff the current CompTIA Data+ objectives, crosswalk every artifact to the five DA0-002 domain IDs, flag any uncovered topics as explicit gaps, and record the next analytics or exam-scheduling decision rather than inferring a pass from coverage.
Retained artifacts
A data dictionary classifying every column of the chosen public dataset by analytical data type, storage type, and structure
A cleaned-and-joined analysis-ready dataset (exported CSV) with a written data-quality log of duplicates, missing values, and join/coverage issues
A descriptive-statistics analysis with grouped aggregates, an outlier flag, a spreadsheet cross-check, and a descriptive-versus-inferential classification
An audience-fit dashboard on public data with fit-for-data charts, an interactive control, a plain-English executive summary, and a completed honest-design review
A governance packet on synthetic data: field-level PII/PHI classification, a data-lineage note, a data-quality health check, and a three-bullet anonymization plan
A five-domain crosswalk with an objectives diff and a note confirming every step used free tools and only public or synthetic data
Review checklist
The concepts, acquisition, analysis, visualization, and governance steps describe one consistent analytics project on a single public dataset (with a synthetic overlay for governance).
Every lab used only public or synthetic data with free tools; no real PII, PHI, financial, regulated, customer, or employer data was used, and nothing sensitive was uploaded to a hosted tool.
Data-quality issues (duplicates, missing values, join grain, coverage gaps) were detected, and each was fixed or flagged transparently rather than hidden.
The statistics are correct and honestly framed: descriptive summaries are not presented as population conclusions, outliers are examined not just dropped, and correlation is not called causation.
The dashboard uses fit-for-data chart types with labeled axes, stated units, and an honest zero baseline, and communicates the takeaway to its stated audience.
The governance packet correctly classifies PII/PHI on synthetic data, includes a lineage note and a quality check, and drafts an anonymization plan, with the data confirmed entirely fabricated.
The current CompTIA Data+ objectives were rechecked and any changed objective invalidates the affected mapping or review.
All five current DA0-002 domains map to at least one artifact; uncovered topics remain explicit gaps rather than implied completion.
The packet does not claim exam success, official CompTIA approval or training beyond linked sources, professional data-analytics experience, authorization to handle real regulated data, or a RoleMath credential.
Safety boundary: Run the entire capstone on free tools with ONLY public or synthetic data. Never use real PII, PHI, financial, regulated, customer, or employer data, and never upload sensitive data to a hosted tool such as Looker Studio - use the local Power BI Desktop and SQLite routes for anything that could be sensitive, and keep the governance step entirely synthetic.
Finish honestly
Completion, portfolio, and maintenance
Completion evidence
All five current DA0-002 domain modules have been covered and checked against the official CompTIA Data+ objectives, including a recheck of the current objectives before any exam scheduling.
Every domain lab has been run with free tools on public or synthetic data only - never real regulated data, and never uploading sensitive data to a hosted tool - and its labeled artifact retained.
Every authored knowledge check has been attempted and each miss has a cited correction plus a fresh scenario.
The CompTIA objectives and the free tool and course resources have been used within their current free-access terms, with any community resource reconciled to the official objectives and its DA0-002 currency verified.
The end-to-end analytics capstone passes its consistency, data-safety, quality, honest-statistics, honest-design, governance, and five-domain coverage review, with the governance step kept synthetic.
The learner has recorded remaining objective gaps and a next analytics or exam-scheduling decision; completion is not represented as an exam result, a credential, professional data-analytics experience, authorization to handle real regulated data, or job readiness.
Portfolio candidates
A data dictionary classifying a public dataset by data type and structure
A cleaned-and-joined analysis-ready dataset with a data-quality log
A descriptive-statistics analysis with a spreadsheet cross-check and an outlier note
An audience-fit dashboard on public data with a plain-English executive summary and a design review
A governance packet on synthetic data: PII/PHI classification, a lineage note, a quality check, and an anonymization plan
The integrated end-to-end capstone packet with a five-domain crosswalk and an objectives diff
Present the packet as self-directed Data+ analytics practice done with free tools on public and synthetic data. Do not call it professional data-analytics experience, CompTIA approval, authorization to handle real regulated data, or a RoleMath credential, and never publish real personal or regulated data.
Freshness controls
Objective source checked 2026-07-11. Recheck objectives every 30 days and resources every 90 days.
Stop and re-verify when
CompTIA changes the Data+ objectives, domain set, weights, exam code, format, passing score, languages, validity/CEU terms, recommended experience, or credential terms.
A free tool (DB Browser for SQLite, LibreOffice Calc, Power BI Desktop, Google Looker Studio) or a public dataset source (UCI, World Happiness, Gapminder) changes URL, access, its free tier, version, or reuse terms.
A community learning resource (SQLBolt, Kaggle Learn, Khan Academy) changes URL, access, or its free tier, or a community practice resource is confirmed to be DA0-001-era rather than DA0-002.
A lab can no longer be run on free tools with public or synthetic data at no cost, or its data-safety guarantee (public/synthetic only; nothing sensitive uploaded to hosted tools; governance synthetic) no longer holds.
A data-concepts, acquisition, analysis, visualization, or governance concept materially changes, or a new topic (for example added AI, cloud, or automation content) is added to or removed from the objectives.
Any module, lab, check, phase, capstone step, account instruction, data-safety guardrail, or objectives diff fails technical, source, analytics-domain, entry-level-framing, data-safety, privacy, accessibility, currency, or claims review.
14%Data governanceCompTIA Data+ (DA0-002) exam objectives and certification page (2026-07-11)
Suggested study order
For Data+ we recommend studying the domains in their published order, because that order already traces the real lifecycle of an analytics project rather than fighting it. CompTIA weights the five domains as Data concepts and environments 20%, Data acquisition and preparation 22%, Data analysis 24%, Visualization and reporting 20%, and Data governance 14%, so the study weight rises toward the analytical core and then tapers - and that curve happens to match the order you should learn in. We open with Data concepts and environments (Domain 1) because it is the vocabulary and mental model everything else assumes: what a data type is, how structured, semi-structured, and unstructured data differ, where data lives (databases, warehouses, lakes, files), and what the common tools do. You cannot prepare data you cannot describe, so this foundation comes first even though it is not the heaviest slice. Data acquisition and preparation (Domain 2, 22%) comes next because in real analytics work most of the effort is here - pulling data from sources, cleaning it, handling duplicates and missing values, joining tables, and shaping it into an analysis-ready form. Getting a trustworthy dataset is the precondition for every honest conclusion downstream. Data analysis (Domain 3, 24%) is the heaviest domain and the natural centerpiece: only once you have clean, joined data do you compute descriptive statistics, run aggregates, distinguish descriptive from inferential reasoning, and find the patterns and outliers that actually answer a question, so it earns the most study time. Visualization and reporting (Domain 4, 20%) follows because an insight nobody can see is not yet useful: here you choose the right chart for the data, build dashboards and reports fit for a specific audience, and decide between static and dynamic outputs. We close with Data governance (Domain 5, 14%) because it is the lightest slice and the one best understood once you have handled real data end to end - classifying sensitive fields like PII and PHI, protecting privacy, keeping data quality high, tracking lineage, and respecting compliance obligations make the most sense after you have seen how data flows through the whole pipeline. In short: understand the data, get it clean, analyze it, show it, and govern it. This is sequencing advice based on the published weights and the analytics lifecycle the exam models, not a claim about the science of learning; if a different order fits how you think, use it.
Study this first. At 20% it is the vocabulary and mental model every other domain assumes: what a data type is, how structured, semi-structured, and unstructured data differ, where data lives, and what the common tools do. You cannot prepare or analyze data you cannot first describe.
This is the 'learn the language of data before you touch any' domain, and CompTIA weights it at 20% of the exam. It is the foundation the other four domains build on: before you acquire, clean, analyze, visualize, or govern data, you have to be able to describe it precisely - its type, its structure, where it is stored, and which tool is right for the job. We study it first because every later domain silently assumes this vocabulary. Data+ is an analytics certification, not a programming one, so the goal here is fluent, correct description and classification, not writing software.
Data types are the first thing to internalize, and the exam expects you to distinguish the analytical category of a value from how a system stores it. Categorical (nominal or ordinal) data labels things - a department name, a satisfaction rating; discrete data counts whole units - number of orders; continuous data measures on a scale - temperature, revenue; boolean data is true/false; and dates and text are their own kinds. The classic trap is treating anything numeric as a number to average: a ZIP code or a product ID is stored as digits but is really categorical, and averaging it is meaningless. Getting types right drives every later choice - which cleaning step, which statistic, which chart.
Data structure is the second axis. Structured data lives in neat rows and columns with a fixed schema - the world of relational databases and spreadsheets. Semi-structured data carries its own tags or keys but no rigid table shape - JSON, XML, log files. Unstructured data has no predefined model at all - free text, images, audio. The exam wants you to recognize which is which and understand the trade-off: structured data is the easiest to query and analyze, while semi-structured and unstructured data are more flexible but need more preparation before analysis. Most entry-level analytics work lives in the structured and semi-structured world.
Data environments - where data actually lives - are the third pillar, and the vocabulary matters. A database (often relational, queried with SQL) stores current operational data. A data warehouse consolidates cleaned, structured data from many systems for reporting and analysis. A data lake holds large volumes of raw data in many formats, structured and unstructured, for flexible later use. A data mart is a focused slice of a warehouse for one team or subject. And plenty of real analytics still starts from flat files - CSVs and spreadsheets. You should be able to match a scenario to the right environment and understand why an analyst pulls from a warehouse for a stable report but from a database or file for ad-hoc work.
Data sources and the common tool landscape round out the domain. Sources include operational databases, APIs, exported files, public open-data portals, and third-party feeds, and each arrives in a different shape and cadence. On the tooling side, the exam expects familiarity with what categories of tool do rather than deep mastery of any one: SQL and database clients for querying, spreadsheets for lightweight analysis, business-intelligence and visualization tools for reporting, and statistical or notebook environments for deeper analysis. The recurring skill is matching a source and a tool to a task, and understanding that the choice of environment shapes what you can later do with the data.
Study this domain by classifying a real public dataset by hand, because correct description is the examinable skill and the foundation of everything after. The lab below has you load the classic UCI Iris dataset into a local SQLite database with the free DB Browser for SQLite, inspect each column with SQLite's typeof() function, and build a data dictionary that records each field's analytical data type, storage type, and structure - turning abstract type-and-structure vocabulary into a concrete artifact you produced. As always, read the official Data+ objectives for CompTIA's authoritative topic list; this explanation paraphrases its scope in our own words rather than reproducing it.
Learn it free
Official · Official exam objectives
CompTIA Data+ (DA0-002) exam objectivesThe exam's own topic list for this domain's data-concepts-and-environments coverage - read it directly rather than relying on any summary, including ours. (captured 2026-07-11)
Vetted independent · Free local tool
DB Browser for SQLiteA free, offline desktop tool for creating a database, importing CSVs, and inspecting column types - the environment this domain's lab uses (verify it is still free before relying on it). (captured 2026-07-11)
Vetted independent · Public dataset source
UCI Machine Learning Repository (public datasets)A free, public source of clean datasets (including Iris) for practicing type and structure classification without any private data. (captured 2026-07-11)
Dataplus Concepts Classification Lab
Classify each column of a real public dataset by analytical data type and structure Distinguish a value's analytical data type from how a database physically stores it
Free tools
Windows, macOS, or Linux with DB Browser for SQLite
Any text or JSON editor to view semi-structured data
Steps
Download the public UCI Iris CSV, open DB Browser for SQLite, create a new local database, and import the CSV as a table named iris.
Run typeof() on a numeric column and the species column to see how SQLite stores each, and confirm species is a small fixed label set.
For each column, record its ANALYTICAL data type (the four measurements are continuous; species is categorical/nominal) alongside its storage type, and note that they are not the same thing.
Fill in the data-dictionary template with one row per column (analytical type, storage type, structure, example, nullability) and add one sentence contrasting the structured table with a semi-structured JSON view of the same rows.
What you should see
Confirm the data dictionary records every Iris column with an analytical data type and a storage type, correctly marks the four measurements continuous and species categorical, and includes a structured-versus-semi-structured note.
Practice evidence maps to exam_domain_comptia_data_plus_da0_002_01
Stay safe & legal: This lab runs entirely on your own machine against the small PUBLIC UCI Iris dataset; never load real PII, PHI, financial, regulated, or employer data into a practice database, and never upload any data to a third-party tool in this lab. Account required: no; payment required: no; maximum designed cost: $0.
Check yourself
2RoleMath-original concept checks for this domain — written by us against cited public sources, never taken from any exam. They confirm understanding; they don’t predict a pass.
Module 2 of 5 · domain 2 · 22% of the exam
Data acquisition and preparation
Study this second, after the concepts foundation. At 22% it is the second-heaviest domain and where most real analytics effort lives: pulling data from sources, cleaning it, handling duplicates and missing values, joining tables, and shaping an analysis-ready dataset. A trustworthy dataset is the precondition for every honest conclusion downstream.
This is the 'get the data, and get it clean' domain, and CompTIA weights it at 22% - the second-heaviest slice of the exam. In real analytics work this is where most of the time goes: the raw data an analyst receives is almost never ready to analyze, and turning messy inputs into a trustworthy, analysis-ready dataset is the daily craft. We study it second because you cannot analyze data you have not first acquired and prepared, and because a single missed duplicate or a wrong join can quietly corrupt every downstream conclusion. This domain is genuinely hands-on, so plan to spend real time in SQL and a spreadsheet.
Acquisition is the first half - getting data from wherever it lives into a workspace you can use. Data arrives from operational databases you query, from APIs you call, from exported files (CSV, Excel, JSON), from public open-data portals, and from third-party feeds. Two related ideas matter here: extraction (pulling the data out) and the broader ETL pattern - extract, transform, load - where data is extracted from sources, transformed into a clean and consistent shape, and loaded into a destination such as a warehouse. The variation ELT loads first and transforms in place; the exam wants you to recognize the pattern and know that acquisition sets the ceiling on data quality - you cannot recover information the source never captured.
Cleaning is the heart of the domain and the most examinable skill. Real data has duplicate rows, missing values, inconsistent formats (dates written five ways, 'NY' versus 'New York'), out-of-range or impossible values, and outright errors. Preparation means detecting and handling each: de-duplicating, deciding how to treat missing values (drop the row, fill with a default or an average, or flag it - each with consequences), standardizing formats and units, correcting or removing invalid values, and validating that the result is internally consistent. The professional discipline the exam rewards is doing this transparently - logging what you changed and why - so the analysis that follows is defensible rather than a black box.
Combining data is the other essential skill, and joins are where beginners most often go wrong. Analysis usually needs data from more than one table, combined on a shared key. An inner join keeps only rows that match in both tables; a left join keeps every row from the left table and fills unmatched right-side columns with nulls, which is exactly how you discover records that failed to match. The subtle trap the exam probes is join grain: if the key is not unique on one side, a join can multiply rows and silently inflate every later count and sum. Understanding one-to-one, one-to-many, and the danger of a many-to-many join is what separates a correct dataset from a plausible-looking wrong one.
Shaping and validation close the domain. Beyond cleaning and joining, preparation often means deriving new columns (a year-over-year change, a category from a range), aggregating to the right grain, filtering to the relevant scope, and pivoting or reshaping between wide and long layouts so the data fits the analysis and the tool. Then you validate: do the row counts make sense, do totals reconcile, are there still nulls where there should not be? The deliverable of this domain is not raw data - it is a documented, analysis-ready dataset whose quality issues you found, fixed, or flagged, so the analysis phase can trust it.
Study this domain by cleaning and joining real public data by hand, because preparation is a doing skill the exam tests directly. The lab below has you import two years of the public World Happiness Report into SQLite with DB Browser, find duplicate and missing values with GROUP BY / HAVING, join the two years on country (using a left join to expose countries that dropped out), derive a year-over-year change column, export the analysis-ready result, and log every quality issue you found. As always, read the official Data+ objectives for CompTIA's authoritative topic list; this explanation paraphrases its scope in our own words rather than reproducing it.
Learn it free
Official · Official exam objectives
CompTIA Data+ (DA0-002) exam objectivesThe authoritative topic list for this domain's acquisition, cleaning, and joining coverage - read it directly rather than relying on any summary, including ours. (captured 2026-07-11)
Vetted independent · Free interactive SQL practice
SQLBolt: interactive SQL lessonsFree, browser-based lessons covering SELECT, filtering, and JOINs - the exact SQL this domain's cleaning and joining work relies on (verify it is still free before relying on it). (captured 2026-07-11)
Vetted independent · Free micro-courses
Kaggle Learn: Data Cleaning and SQL micro-coursesFree short courses on data cleaning, Pandas, and SQL that reinforce the preparation skills this domain centers on (a free account is required; verify the free label before relying on it). (captured 2026-07-11)
Dataplus Acquisition Prep Lab
Detect and handle duplicates and missing values in a real public dataset with SQL Join two tables correctly, expose unmatched rows with a left join, and derive an analysis-ready column
Free tools
Windows, macOS, or Linux with DB Browser for SQLite
A spreadsheet application to inspect the exported CSV (optional)
Steps
Import two years of the public World Happiness Report into SQLite as two tables and confirm their row counts.
Run the duplicate and missing-value checks and record what you find as data-quality issues.
Left-join the two years on country to expose dropped countries, then inner-join to derive a year-over-year change column.
Export the derived, analysis-ready result to CSV and complete the data-quality log with the duplicates, missing values, and dropped countries you handled.
What you should see
Confirm the learner ran duplicate and missing-value checks, used a left join to expose unmatched countries, derived a year-over-year column, exported an analysis-ready CSV, and logged each quality issue with a handling decision.
Practice evidence maps to exam_domain_comptia_data_plus_da0_002_02
Stay safe & legal: This lab uses only the PUBLIC World Happiness Report on your own machine; never substitute real employer, customer, personal, financial, or regulated data into these tables, and never upload the database or its data to any third-party tool. Account required: no; payment required: no; maximum designed cost: $0.
Check yourself
2RoleMath-original concept checks for this domain — written by us against cited public sources, never taken from any exam. They confirm understanding; they don’t predict a pass.
Module 3 of 5 · domain 3 · 24% of the exam
Data analysis
Study this third, once you can produce a clean, joined dataset. At 24% it is the heaviest domain and the analytical centerpiece: descriptive statistics, SQL aggregates, the difference between describing your data and inferring about a wider population, and finding the patterns and outliers that answer a question.
This is the 'turn a clean dataset into an answer' domain, and CompTIA weights it at 24% - the heaviest single slice of the exam. Everything before this got you a trustworthy dataset; this domain is where you actually analyze it: summarize it with statistics, aggregate it with SQL, compare groups, spot patterns and outliers, and reason honestly about what the numbers do and do not support. We study it third because it depends entirely on the clean, joined data from the preparation domain, and we give it the most study time because it is both the largest domain and the intellectual core of an analyst's job. It is not a data-science exam, so the statistics are foundational, not advanced modeling.
Descriptive statistics are the starting point and the most examinable content. These describe the data you actually have: measures of central tendency (mean, median, mode) tell you the typical value, and measures of spread (range, variance, standard deviation) tell you how much the values vary. The exam expects real judgment here - the mean is pulled around by outliers while the median is resistant, so a mean far from the median is itself a signal of skew or extreme values. Knowing which summary to report for a given distribution, and reading a standard deviation as 'how tightly the data clusters,' is a core Data+ skill.
The descriptive-versus-inferential distinction is a concept the exam tests directly and one analysts must respect. Descriptive statistics summarize the dataset in front of you and make no claim beyond it. Inferential statistics use a sample to make a claim about a larger population you did not fully measure - hypothesis tests, confidence intervals, correlation, and regression live here. The professional discipline is not over-claiming: if you measured every record, you are describing; if you measured a sample and want to generalize, you are inferring, and that carries uncertainty. Data+ keeps inferential statistics at a conceptual, foundational level, but expects you to know which mode of reasoning a situation calls for.
SQL aggregation is the hands-on engine of this domain. Aggregate functions - COUNT, SUM, AVG, MIN, MAX - collapse many rows into a summary, and GROUP BY computes those summaries per category, which is how an analyst compares segments (average score per continent, total sales per region). HAVING filters those grouped results. The exam expects you to read and reason about aggregate queries and to understand what a grouped result means. Because some tools (like plain SQLite) lack a built-in median or standard deviation, part of the skill is knowing when to compute a statistic in SQL versus a spreadsheet, and cross-checking that the two agree.
Pattern-finding and outlier analysis close the analytical work. Beyond single summaries, analysis means looking for trends over time, relationships between variables (does one move with another?), and anomalies - values so far from the rest that they deserve investigation. A common, defensible rule of thumb flags points more than two standard deviations from the mean as potential outliers, but the judgment is deciding whether an outlier is an error to fix, a genuine extreme to keep, or the very thing the question is about. Correlation is introduced here with the standing caution that it does not prove causation. The deliverable of this domain is a defensible answer: the statistics that support it, the outliers you examined, and an honest note on what the data does not let you conclude.
Study this domain by computing descriptive statistics and SQL aggregates on real public data and cross-checking them, because that dual-tool discipline is exactly the skill. The lab below has you load a Gapminder-style public dataset into SQLite, compute count/min/max/range/mean and per-continent GROUP BY aggregates, flag outliers beyond two standard deviations with a subquery, then cross-check the mean, median, and standard deviation in a spreadsheet and classify each finding as descriptive or inferential. As always, read the official Data+ objectives for CompTIA's authoritative topic list; this explanation paraphrases its scope in our own words rather than reproducing it.
Learn it free
Official · Official exam objectives
CompTIA Data+ (DA0-002) exam objectivesThe authoritative topic list for this domain's descriptive-statistics, aggregation, and analysis coverage - read it directly rather than relying on any summary, including ours. (captured 2026-07-11)
Vetted independent · Free statistics course
Khan Academy: Statistics and ProbabilityA free, nonprofit course covering central tendency, spread, distributions, and the descriptive-versus-inferential distinction this domain rests on (verify it is still free before relying on it). (captured 2026-07-11)
Vetted independent · Free micro-courses
Kaggle Learn: Intro to SQL and Pandas micro-coursesFree short courses on SQL aggregation and data analysis that reinforce the GROUP BY and summary-statistic skills this domain tests (a free account is required; verify the free label before relying on it). (captured 2026-07-11)
Dataplus Analysis Stats Lab
Compute descriptive statistics and grouped SQL aggregates on a real public dataset Flag outliers with a standard-deviation rule and classify findings as descriptive or inferential
Free tools
Windows, macOS, or Linux with DB Browser for SQLite
LibreOffice Calc offline or a free online spreadsheet for the cross-check
Steps
Import the public Gapminder-style CSV into SQLite and compute overall descriptive statistics (count, min, max, range, mean) for the numeric measure.
Compute a grouped aggregate to compare segments (mean measure per continent or region).
Use the fixture's subquery to compute the mean and standard deviation and return rows more than two standard deviations from the mean.
Cross-check the mean, median, and standard deviation in a spreadsheet, note any mean-versus-median skew, and classify each finding as descriptive or inferential.
What you should see
Confirm the learner computed descriptive statistics and a grouped aggregate in SQL, flagged outliers beyond two standard deviations, reconciled the mean with a spreadsheet, and classified each finding as descriptive or inferential.
Practice evidence maps to exam_domain_comptia_data_plus_da0_002_03
Stay safe & legal: This lab uses only PUBLIC Gapminder-style data on your own machine; if you use an online spreadsheet, place only public data in it, and never paste real employer, customer, personal, financial, or regulated data into any spreadsheet or SQL tool. Account required: optional; payment required: no; maximum designed cost: $0.
Check yourself
2RoleMath-original concept checks for this domain — written by us against cited public sources, never taken from any exam. They confirm understanding; they don’t predict a pass.
Module 4 of 5 · domain 4 · 20% of the exam
Visualization and reporting
Study this fourth, after analysis. At 20% it is where insight becomes useful: choosing the right chart for the data, building dashboards and reports fit for a specific audience, and deciding between static and dynamic outputs. An insight nobody can see is not yet useful.
This is the 'make the insight visible and understandable' domain, and CompTIA weights it at 20% of the exam. Analysis produced an answer; this domain turns that answer into something a decision-maker can actually see and act on. The core skill is not making pretty pictures - it is communication: choosing a visual that fits the data and the question, and building a report or dashboard fit for a specific audience. We study it fourth because you cannot visualize a conclusion you have not yet reached, and because good visualization depends on understanding the data types and statistics from the earlier domains.
Choosing the right chart for the data is the most examinable skill, and the exam expects real judgment. Match the visual to the message: a line chart shows a trend over time; a bar or column chart compares values across categories; a pie or stacked chart shows composition (parts of a whole) and only works with a few categories; a histogram shows the distribution of a single variable; a scatter plot shows the relationship between two variables; and a single scorecard or KPI highlights one key number. The classic mistakes the exam probes are using a pie chart for a trend, a line chart for unordered categories, or a 3D effect that distorts values - the chart should reveal the data, not decorate it.
Honest, readable design is the second pillar. A visualization can be technically correct and still mislead: a bar chart with a truncated y-axis exaggerates differences, an unlabeled axis leaves the reader guessing, missing units make numbers meaningless, and poor color choices hide the point or fail color-blind viewers. The exam rewards knowing the guardrails - start bar charts at zero, label axes and state units, use a legend where color encodes meaning, and keep the design in service of the message. Clarity beats cleverness: a plain bar chart with an honest axis communicates better than a crowded dashboard of dials.
Dashboards and reporting are where multiple visuals come together, and audience is the organizing idea. A dashboard combines several charts, a headline number, and often interactive controls into one view of a topic. The exam expects you to tailor the output to who will read it: an executive wants a few clear headline numbers and a one-line takeaway; an analyst wants detail and the ability to drill down; an operations team wants live status. A good report or dashboard answers its audience's question without a manual, leads with the takeaway, and does not bury the signal under decoration.
Static versus dynamic and the reporting cadence round out the domain. A static report is a fixed snapshot - a PDF or an image - suited to a formal deliverable or a point-in-time summary. A dynamic dashboard is interactive and often connected to live or refreshed data, letting viewers filter, change date ranges, and explore. Related choices include how often a report refreshes and whether it is scheduled or on-demand. The exam wants you to pick the right mode for the purpose: a board deck is static; a daily operations monitor is dynamic. The through-line of the whole domain is fit-for-audience communication, not chart count.
Study this domain by building a real dashboard on public data and reviewing it against a communication checklist, because fit-for-audience design is the examinable skill. The lab below has you connect a PUBLIC CSV to a free BI tool - Google Looker Studio (or Power BI Desktop if you prefer local) - build a time-series chart, a sorted bar chart, a scorecard, and an interactive filter, write a plain-English executive summary, and then review the result against a chart-choice and honest-design checklist. Because Looker Studio is a hosted tool, you connect only PUBLIC data - never real employer or personal data. As always, read the official Data+ objectives for CompTIA's authoritative topic list; this explanation paraphrases its scope in our own words rather than reproducing it.
Learn it free
Official · Official exam objectives
CompTIA Data+ (DA0-002) exam objectivesThe authoritative topic list for this domain's visualization and reporting coverage - read it directly rather than relying on any summary, including ours. (captured 2026-07-11)
Vetted independent · Free BI tool
Google Looker StudioA free, hosted business-intelligence tool for building dashboards and reports on public data - the environment this domain's lab uses (a free Google account is required; connect public data only). (captured 2026-07-11)
Vetted independent · Free micro-course
Kaggle Learn: Data Visualization micro-courseA free short course on choosing and building charts that reinforces the chart-selection judgment this domain tests (a free account is required; verify the free label before relying on it). (captured 2026-07-11)
Dataplus Visualization Dashboard Lab
Build an audience-fit dashboard with the right chart types on public data in a free BI tool Review a dashboard against an honest-design and chart-choice checklist
Free tools
Google Looker Studio in a browser (hosted, free account)
Power BI Desktop on Windows (local, no account) as the offline alternative
Steps
Complete the dashboard-spec checklist's audience, purpose, the one question it answers, and the static-versus-dynamic decision before building.
Connect a PUBLIC CSV to Looker Studio (or import it into Power BI Desktop), first confirming it contains no PII, PHI, or employer data.
Build a labeled time-series line chart, a value-sorted zero-based bar chart, a scorecard for the key number, and an interactive filter or date-range control.
Write a plain-English executive summary of the takeaway, then complete the design-review checklist (labeled axes, units, legend, honest baseline, chart-type fit).
What you should see
Confirm the dashboard uses fit-for-data chart types (time series, sorted zero-based bar, scorecard), includes an interactive control, carries a plain-English executive summary, and passes the honest-design review, all on public data.
Practice evidence maps to exam_domain_comptia_data_plus_da0_002_04
Stay safe & legal: Looker Studio is a hosted third-party tool, so connect ONLY public or synthetic data to it; never upload real PII, PHI, financial, customer, or employer data to Looker Studio or any hosted BI service - use the local Power BI Desktop route for anything sensitive. Account required: optional; payment required: no; maximum designed cost: $0.
Check yourself
2RoleMath-original concept checks for this domain — written by us against cited public sources, never taken from any exam. They confirm understanding; they don’t predict a pass.
Module 5 of 5 · domain 5 · 14% of the exam
Data governance
Study this last. At 14% it is the lightest slice and the one best understood after you have handled real data end to end: classifying sensitive fields like PII and PHI, protecting privacy, keeping data quality high, tracking lineage, and respecting compliance obligations.
This is the 'handle data responsibly and legally' domain, and CompTIA weights it at 14% - the lightest of the five. Governance is the set of rules, roles, and practices that keep data accurate, secure, private, and compliant across its whole life. We study it last because it makes the most sense once you have seen data flow through the entire pipeline: you cannot govern what you have not yet acquired, analyzed, and reported on. It is the smallest domain by weight but a load-bearing one professionally - a single mishandled sensitive field can turn a good analysis into a privacy incident, so the discipline here protects both people and the organization.
Classifying data by sensitivity is the most examinable skill, and the acronyms matter. Personally identifiable information (PII) is any data that can identify a specific person - name, email, government ID, address, and often combinations that become identifying together. Protected health information (PHI) is health data tied to an individual, which carries extra legal weight. Financial and payment data (for example cardholder data) is its own sensitive class. The exam expects you to look at a dataset and correctly flag which fields are sensitive and at what level, because that classification drives every protection decision that follows - who may see it, how it is stored, and whether it may leave the organization.
Privacy protection and the major regulations form the second pillar, kept at a conceptual level for Data+. You should recognize the landscape: the GDPR governs personal data of people in the EU; HIPAA governs protected health information in US healthcare; and other rules (such as CCPA) govern specific regions or sectors. Just as important are the techniques that reduce risk - anonymization (irreversibly removing identifiers), pseudonymization (replacing identifiers with tokens that can be mapped back only with a separate key), masking, and aggregation. The principle the exam rewards is data minimization: collect and expose only what the task genuinely needs, and de-identify wherever possible.
Data quality and its dimensions are the third pillar, and they connect governance back to trustworthy analysis. Governance defines and monitors quality across dimensions like accuracy (is it correct?), completeness (are values missing?), consistency (does it agree across systems?), timeliness (is it current?), uniqueness (are there duplicates?), and validity (does it conform to the rules?). This is where the cleaning work of Domain 2 gets a governance frame: quality is not a one-time cleanup but an ongoing, measured commitment, often expressed as data-quality rules and health checks. Poor-quality data governed badly produces confident, wrong conclusions.
Lineage, roles, and the data lifecycle close the domain. Data lineage traces where data came from, how it was transformed, and where it flows - essential for trusting a number and for auditing it if someone challenges it. Governance also assigns roles: data owners are accountable for a data domain, data stewards maintain its quality and definitions, and custodians manage the technical storage. And it spans the lifecycle from creation and storage through use, archival, and secure disposal, including retention rules that say how long data may or must be kept. The exam wants you to understand governance as an ongoing, role-based, lifecycle-wide practice, not a single control.
Study this domain by building a governance packet for a dataset you can classify freely, and this is the one lab where the data MUST be synthetic. The lab below has you create (or use the provided) synthetic 20-row fake patient-appointment dataset with obviously-fake names, build a data dictionary that classifies each field's sensitivity as public, PII, or PHI, write a short data-lineage note and a data-quality health check, and draft a three-bullet anonymization plan - all on entirely fabricated data, because you must never practice governance on real regulated records. As always, read the official Data+ objectives for CompTIA's authoritative topic list; this explanation paraphrases its scope in our own words rather than reproducing it.
Learn it free
Official · Official exam objectives
CompTIA Data+ (DA0-002) exam objectivesThe authoritative topic list for this domain's data-governance coverage - read it directly rather than relying on any summary, including ours. (captured 2026-07-11)
Vetted independent · Free micro-course
Kaggle Learn: Data Ethics / Intro to AI Ethics micro-courseA free short course touching data privacy, bias, and responsible data handling that reinforces the governance mindset this domain tests (a free account is required; verify the free label before relying on it). (captured 2026-07-11)
Vetted independent · Free local tool
DB Browser for SQLiteA free, offline tool to inspect the synthetic governance dataset as a table if you prefer SQL over a spreadsheet - kept local so no data leaves your machine (verify it is still free before relying on it). (captured 2026-07-11)
Dataplus Governance Classification Lab
Classify each field of a synthetic dataset as public, PII, or PHI and justify the level Produce a governance packet: a lineage note, a data-quality health check, and an anonymization plan on synthetic data
Free tools
Any operating system with a text editor or spreadsheet
DB Browser for SQLite optionally, to inspect the synthetic CSV as a table
Steps
Open the synthetic patient-appointment dataset and confirm every name is obviously fake and every row is labeled synthetic before classifying anything.
Classify each field's sensitivity in the data dictionary: name and email as PII, date_of_birth and record_number as identifying PII, diagnosis_code and department as PHI, and appointment_id as a non-sensitive key.
Write a short data-lineage note (source, transformations, downstream use) and run a data-quality health check across completeness, uniqueness, and validity.
Draft a three-bullet anonymization plan (for example drop/hash name and email, generalize date_of_birth to a year or band, tokenize record_number) and note conceptually which regulation each sensitive class implicates.
What you should see
Confirm the learner classified every field as public/PII/PHI, correctly marked the health fields PHI, wrote a lineage note, ran a multi-dimension quality check, and drafted an anonymization plan, all on entirely synthetic data.
Practice evidence maps to exam_domain_comptia_data_plus_da0_002_05
Stay safe & legal: This lab must use ENTIRELY SYNTHETIC data - fabricated names, fake emails, invented record numbers - on your own machine; never use, download, import, or practice governance on real patient, customer, employee, financial, or any regulated data, and never upload even the synthetic set to a hosted tool as a real-data template. Account required: no; payment required: no; maximum designed cost: $0.
Check yourself
2RoleMath-original concept checks for this domain — written by us against cited public sources, never taken from any exam. They confirm understanding; they don’t predict a pass.
Before you book the exam
Work through the modules above, then get a personalized read on where you stand: the readiness check maps your background against these same published domains and suggests what to study first — no score, no pass prediction.
Exam registration fee: Approximately $264 USD. CompTIA prices exams by country or region and does not print a dollar figure in the exam objectives PDF, so treat roughly $264 as an approximate US voucher price and confirm the exact current fee for your location on the official CompTIA store (store.comptia.org) before you register. Official CompTIA Data+ (DA0-002) certification page (region-priced; no figure printed in the objectives)
Languages and validity: Delivered in English and Japanese. The certification is valid for three years and can be renewed through CompTIA's continuing-education (CE) program; the exact number of continuing-education units required is set by CompTIA's CE program (commonly 20 CEUs at this tier) rather than printed on the certification page, so confirm the current requirement on CompTIA's CE program pages. Official CompTIA Data+ (DA0-002) certification page (validity; CEU count set by the CE program, not printed here)
Recommended experience and level: An entry-to-early-career data analytics certification. CompTIA recommends roughly 18-24 months of hands-on experience with data or business analytics; it is not a data-science, data-engineering, or programming credential. No programming is required, though spreadsheet literacy is assumed and SQL plus basic statistics fluency shorten preparation. Official CompTIA Data+ (DA0-002) certification page
Version currency: DA0-002 (V2) is the current version, launched October 14, 2025; the prior DA0-001 version retired April 14, 2026 (English). Always diff the current objectives on the official page before you schedule. Official CompTIA Data+ (DA0-002) certification page
A free, source-cited study companion built on CompTIA's published Data+ (DA0-002) exam objectives, for independent study only. It is not official training, is not affiliated with or endorsed by CompTIA, and is not a pass guarantee. Data+ is an entry-to-early-career data analytics certification - CompTIA recommends roughly 18-24 months of hands-on data or business-analytics experience - not a data-science, data-engineering, or programming credential; no programming is required, though spreadsheet literacy is assumed and SQL and statistics fluency shorten preparation. Every hands-on lab uses free tools and only PUBLIC or SYNTHETIC data with no cloud cost, and you must never upload real PII, PHI, financial, regulated, or employer data to any third-party tool. Verify the current objectives on the official page before your exam.
Certification and vendor names are used only to identify the program this independent study companion refers to. RoleMath is not affiliated with, endorsed by, or sponsored by CompTIA.