I85PREP
⌘ K
CompTIA Data+ study materials
CompTIA Data+ study resources

CompTIA Data+ practice questions

One question at a time. Build your CompTIA Data+ understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Data concepts and environments

Which key is chosen as the main identifier for a table, uniquely identifies every row, and cannot contain null values?

  1. Primary key Correct answer.
  2. Foreign key
  3. Superkey
  4. Nonclustered index

Explanation

A primary key is the column or set of columns designated as the table's main row identifier and must be unique and non-null. Other keys may enforce relationships or improve lookup performance but do not serve this identifying role.

Question 2Data concepts and environments

A company records the number of customer support tickets opened each day. Which classification best describes this value?

  1. Ordinal categorical
  2. Discrete quantitative Correct answer.
  3. Continuous quantitative
  4. Nominal categorical

Explanation

Counts are quantitative because they are numeric measurements, and they are discrete because they can take only whole-number values. Continuous values would be measurable on a scale, such as the average handling time.

Question 3Data concepts and environments

Which item is an example of unstructured data?

  1. An XML invoice with tagged elements
  2. Security camera video files Correct answer.
  3. A CSV table of monthly sales figures
  4. A JSON payload with nested fields

Explanation

Unstructured data lacks a predefined tabular or tagged model, and video is a raw media format. CSV, JSON, and XML all impose some organizing structure on their contents.

Question 4Data concepts and environments

Which file format stores tabular data as plain text with values separated by a delimiter such as a comma and normally carries no embedded schema?

  1. CSV Correct answer.
  2. Parquet
  3. Avro
  4. XML

Explanation

CSV is a delimited text format that represents rows as lines and fields as delimited values, without embedded data types or schema. Parquet and Avro embed schema information, while XML uses tags rather than a simple delimiter.

Question 5Data concepts and environments

In a star schema, which type of table stores numeric measures such as sales amount and quantity at a defined level of detail?

  1. Fact table Correct answer.
  2. Bridge table
  3. Dimension table
  4. Lookup table

Explanation

A fact table holds measurable business events and foreign keys to related dimensions, with each row representing a defined grain. Dimension tables provide descriptive context such as product or customer attributes.

Question 6Data concepts and environments

A customer satisfaction rating recorded as low, medium, or high is what type of data?

  1. Nominal
  2. Interval
  3. Ordinal Correct answer.
  4. Ratio

Explanation

Ordinal data consists of categories with a meaningful order but without a consistent numeric distance between them. Nominal categories lack order, while interval and ratio data are numeric.

Question 7Data concepts and environments

A relational database organizes data primarily into which structure?

  1. Tables of rows and columns Correct answer.
  2. Documents of nested objects
  3. Graphs of nodes and edges
  4. Schema-free key-value pairs

Explanation

Relational databases store data in tables made up of rows and columns, with relationships enforced through keys. Documents, key-value pairs, and graphs are associated with NoSQL database models.

Question 8Data concepts and environments

Which environment typically stores raw data in its native format until it is needed for analysis?

  1. Data warehouse
  2. Data mart
  3. Operational data store
  4. Data lake Correct answer.

Explanation

A data lake holds large volumes of raw data in native formats, including structured, semi-structured, and unstructured files. A warehouse stores modeled and transformed data, while a mart is a focused subset.

Question 9Data concepts and environments

Which format is semi-structured and represents data as nested key-value pairs commonly exchanged through web APIs?

  1. JSON Correct answer.
  2. CSV
  3. Parquet
  4. Fixed-width text

Explanation

JSON uses named fields and nested objects or arrays, giving it a flexible semi-structured layout. CSV and fixed-width text are flat structured formats, and Parquet is a binary columnar format with an embedded schema.

Question 10Data concepts and environments

In a snowflake schema, dimension tables are typically what?

  1. Combined into a single wide table
  2. Normalized into related tables Correct answer.
  3. Stored as unstructured documents
  4. Replaced by fact tables

Explanation

A snowflake schema normalizes dimensions into multiple related tables, reducing redundancy at the cost of additional joins. A star schema keeps dimensions denormalized in single wide tables.

Question 11Data concepts and environments

Which key in a child table references the primary key of a parent table to enforce referential integrity?

  1. Composite key
  2. Candidate key
  3. Surrogate key
  4. Foreign key Correct answer.

Explanation

A foreign key contains values that must match a primary or unique key in a parent table, preventing orphaned child records. Composite, surrogate, and candidate keys describe other ways of identifying rows.

Question 12Data concepts and environments

A table tracks student enrollments and uses both StudentID and CourseID together to uniquely identify each row. What is this an example of?

  1. Surrogate key
  2. Alternate key
  3. Foreign key
  4. Composite key Correct answer.

Explanation

A composite key combines two or more columns to uniquely identify a row, as StudentID and CourseID do for an enrollment. A surrogate key would be a single generated identifier instead.

Question 13Data concepts and environments

A data mart is best described as what?

  1. A repository for raw unprocessed source files
  2. A transactional system for daily order entry
  3. A visualization layer for executive dashboards
  4. A subject-specific warehouse subset Correct answer.

Explanation

A data mart is a smaller, subject-oriented subset of a data warehouse, often built for a department or specific reporting need. Raw files belong in a data lake, and transactional order entry is an OLTP workload.

Question 14Data concepts and environments

Which workload is characterized by complex read queries that scan large volumes of historical data for reporting?

  1. Master data management
  2. OLAP Correct answer.
  3. Data ingestion
  4. OLTP

Explanation

OLAP systems are optimized for analytical queries that aggregate and scan large historical datasets. OLTP systems instead handle many short read and write transactions.

Question 15Data concepts and environments

Which file format stores data by column rather than by row and is optimized for analytical queries?

  1. XML
  2. CSV
  3. Parquet Correct answer.
  4. JSON

Explanation

Parquet is a columnar binary format that stores values of the same column together, enabling efficient compression and selective column reads. CSV, JSON, and XML are row-oriented or text-based formats.

Question 16Data concepts and environments

Which format uses opening and closing tags to describe hierarchical data and is commonly used for data exchange?

  1. Avro
  2. XML Correct answer.
  3. Parquet
  4. CSV

Explanation

XML structures data with nested elements enclosed in tags, allowing hierarchical representation. CSV is delimiter-based, while Parquet and Avro are binary formats with schemas.

Question 17Data concepts and environments

What is the primary goal of normalizing a relational database?

  1. Reduce redundancy and update anomalies Correct answer.
  2. Increase duplicate records for faster reads
  3. Improve dashboard color consistency
  4. Store all attributes in one flat table

Explanation

Normalization organizes attributes into related tables so that each fact is stored once, reducing redundancy and the anomalies that occur during inserts, updates, and deletes. Denormalization intentionally adds redundancy to speed up reads.

Question 18Data concepts and environments

In an ELT process, what happens before data is transformed?

  1. Data is visualized in a report
  2. Data is deleted from the source
  3. Data is loaded into the target system Correct answer.
  4. Data is transformed in the source system

Explanation

ELT stands for extract, load, transform, so data is first loaded into the target system and transformed there using its processing power. ETL instead transforms data before loading it.

Question 19Data concepts and environments

A dataset lists one row per store per month with columns StoreID, Month, and Sales. Which layout is this?

  1. Long format Correct answer.
  2. Wide format
  3. Snowflake format
  4. Denormalized format

Explanation

Long format stores one observation per row, with an identifier column, a variable column, and a value column. Wide format would place each month in a separate column.

Question 20Data concepts and environments

Which NoSQL database type stores data as JSON-like documents with flexible schemas?

  1. Key-value
  2. Relational
  3. Document Correct answer.
  4. Graph

Explanation

Document databases store records as self-contained documents, often JSON-like, and allow each document to vary in structure. Key-value stores use simple key-to-value mappings, and graph databases emphasize relationships.

Question 21Data concepts and environments

A surrogate key is best described as what?

  1. A natural key derived from source data
  2. A generated key with no business meaning Correct answer.
  3. A key that references another table
  4. A combination of all business attributes

Explanation

A surrogate key is created by the system, often as an incrementing integer or UUID, and carries no business meaning. It is stable even when business attributes change, unlike a natural key.

Question 22Data concepts and environments

A fact table contains one row per individual order line. What does this describe?

  1. The schema type
  2. The grain of the fact table Correct answer.
  3. The normalization level
  4. The cardinality of a dimension

Explanation

The grain defines what a single fact table row represents, such as one order line or one daily store total. Declaring the grain before adding measures prevents double counting.

Question 23Data concepts and environments

Compared with a star schema, a snowflake schema generally does what?

  1. Reduces redundancy but requires more joins Correct answer.
  2. Stores data only in flat files
  3. Increases query simplicity and reduces the number of joins
  4. Eliminates the need for dimension tables

Explanation

Snowflake designs normalize dimensions into related sub-tables, which reduces duplicated descriptive data but increases the number of joins needed. Star schemas denormalize dimensions to keep queries simpler and often faster.

Question 24Data concepts and environments

Why does columnar storage often outperform row storage for analytical queries?

  1. It stores each row contiguously for fast inserts
  2. It enforces referential integrity at the storage layer
  3. It reads only the columns referenced by the query Correct answer.
  4. It eliminates the need for a schema

Explanation

Analytical queries usually reference a subset of columns, and columnar storage lets the engine read just those columns instead of entire rows. Storing similar values together also improves compression and scan efficiency.

Question 25Data concepts and environments

A data lakehouse combines which two capabilities?

  1. Flexible lake storage with warehouse management Correct answer.
  2. Data mart reporting with spreadsheet formulas
  3. CSV flat files with XML message queues
  4. OLTP transaction processing with graph traversal

Explanation

A lakehouse uses low-cost, flexible data lake storage while adding warehouse-style features such as schema enforcement, transactions, and governance. This lets one platform support both raw data and curated analytical workloads.

Question 26Data concepts and environments

A dataset stores delivery weights in kilograms, and a weight of 0 kg means the package is genuinely empty. Which measurement scale applies to this column?

  1. Interval
  2. Ordinal
  3. Nominal
  4. Ratio Correct answer.

Explanation

A ratio scale has equal intervals plus a true zero, so statements such as '20 kg is twice 10 kg' are meaningful. Interval scales have equal intervals but no true zero, ordinal scales only rank order, and nominal scales are unordered labels.

Question 27Data concepts and environments

Which item is an example of structured data?

  1. Free-text customer reviews
  2. Audio recordings of sales calls
  3. Relational employee table rows Correct answer.
  4. Scanned invoice images

Explanation

Structured data conforms to a predefined schema of defined columns and types, as found in relational tables. Audio, images, and free-text reviews lack a fixed tabular schema and are unstructured.

Question 28Data concepts and environments

Which NoSQL database type is best suited to store friend-of-a-friend relationships and traverse them quickly?

  1. Document store
  2. Graph database Correct answer.
  3. Key-value store
  4. Wide-column store

Explanation

A graph database stores entities as nodes and relationships as edges, so traversal across connected records is a native operation. Document, key-value, and wide-column stores are optimized for whole-record retrieval rather than relationship traversal.

Question 29Data concepts and environments

Which level of data model describes business entities and their relationships at the highest level of abstraction, independent of any technology?

  1. Dimensional
  2. Conceptual Correct answer.
  3. Logical
  4. Physical

Explanation

A conceptual model captures the business entities and relationships without attributes, keys, or platform details. A logical model adds attributes and keys, and a physical model specifies tables, data types, and storage for a particular platform.

Question 30Data concepts and environments

A table stores multiple phone numbers for a customer as a comma-separated list in a single PhoneNumbers column. Which normal form does this violate?

  1. First normal form Correct answer.
  2. Fourth normal form
  3. Second normal form
  4. Third normal form

Explanation

First normal form requires each column to hold a single atomic value, and a comma-separated list is a repeating group. Second normal form removes partial dependencies on part of a composite key, and third normal form removes transitive dependencies.

Question 31Data concepts and environments

In a dimensional model, what does the grain of a fact table define?

  1. The level of detail represented by one fact row Correct answer.
  2. The sort order of rows when the table is loaded
  3. The number of dimension tables joined to the fact table
  4. The data type used for the numeric measures

Explanation

The grain states exactly what a single fact row represents, such as one sale line or one store per day, and every measure must be consistent with that level. Choosing the grain first prevents mixing measures at different levels of detail in one table.

Question 32Data concepts and environments

How do transactional (OLTP) and analytical (OLAP) systems typically differ in schema design?

  1. Both are equally normalized and differ only in storage engine
  2. OLTP is normalized for integrity; OLAP is denormalized for speed Correct answer.
  3. Both use identical schemas and differ only in backup frequency
  4. OLTP is denormalized for speed; OLAP is normalized for integrity

Explanation

OLTP systems process many small writes and are normalized to avoid update anomalies and redundant storage. OLAP systems run large read queries, so they use denormalized star or snowflake designs that reduce joins and speed aggregation.

Question 33Data concepts and environments

A data catalog records each dataset's source system, file format, and column definitions. Which category of metadata is this?

  1. Technical metadata Correct answer.
  2. Business metadata
  3. Reference metadata
  4. Operational metadata

Explanation

Technical metadata describes structure and storage details such as schema, data types, file format, and source system. Business metadata instead captures definitions, ownership, and usage rules in business terms.

Question 34Data concepts and environments

Which term describes the shared, authoritative data about core business entities such as customers, products, and suppliers that must stay consistent across systems?

  1. Master data Correct answer.
  2. Reference data
  3. Metadata
  4. Transactional data

Explanation

Master data is the single trusted record for core entities shared across applications, often called the golden record. Reference data is a controlled set of allowable values, such as country codes or status codes, used to classify other data.

Question 35Data concepts and environments

A validation rule rejects any record in which a US ZIP code field does not contain exactly five digits. Which data quality dimension does this rule enforce?

  1. Timeliness
  2. Uniqueness
  3. Completeness
  4. Validity Correct answer.

Explanation

Validity measures whether values conform to an accepted format, range, or rule, which is exactly what a five-digit pattern check enforces. Completeness concerns missing values, uniqueness concerns duplicates, and timeliness concerns how current the data is.

Question 36Data concepts and environments

A warehouse assigns each customer an auto-incremented integer identifier that has no meaning in the source system. What is this key called?

  1. Foreign key
  2. Natural key
  3. Composite key
  4. Surrogate key Correct answer.

Explanation

A surrogate key is system-generated, has no business meaning, and insulates the warehouse from changes to source identifiers. A natural key comes from the business data itself, a composite key combines multiple columns, and a foreign key references another table's key.

Question 37Data concepts and environments

A data lake stores raw files without validating their structure at load time and imposes structure only when the data is queried. Which approach does this describe?

  1. Schema-on-write
  2. Snowflake schema
  3. Star schema
  4. Schema-on-read Correct answer.

Explanation

Schema-on-read defers parsing and validation to query time, which suits lakes holding raw files of varying structure. Schema-on-write validates data against a defined schema at load time, as relational warehouses typically do.

Question 38Data concepts and environments

A table has OrderID as its key with non-key columns CustomerID, CustomerZip, and CustomerCity, where CustomerZip determines CustomerCity. Which normal form does this dependency violate?

  1. Third normal form Correct answer.
  2. Domain-key normal form
  3. Second normal form
  4. First normal form

Explanation

Third normal form requires every non-key attribute to depend on the key and nothing else; a non-key attribute depending on another non-key attribute is a transitive dependency. Moving CustomerZip and CustomerCity to their own table removes it.

Question 39Data concepts and environments

A customer dimension must retain the full history of address changes so each fact row joins to the address that was in effect at the time of the sale. Which technique fits this requirement?

  1. Truncate and reload the dimension nightly
  2. Type 2 with new rows and effective dates Correct answer.
  3. Type 1 overwrite of the existing row
  4. Type 3 with one prior-value column

Explanation

A Type 2 slowly changing dimension inserts a new versioned row with effective and expiration dates, preserving full history so historical facts keep pointing at the correct version. Type 1 overwrites history, and Type 3 keeps only a limited number of prior values.

Question 40Data concepts and environments

In a dimensional model, patients can have many diagnoses and each diagnosis applies to many patients. Which technique resolves this many-to-many relationship between the two dimensions?

  1. A bridge table placed between the dimensions Correct answer.
  2. A role-playing dimension used twice in the fact table
  3. A junk dimension combining both dimension keys
  4. A degenerate dimension stored in the fact table

Explanation

A bridge table sits between the two dimensions and holds one row per pairing, often with weighting factors, so multivalued relationships can be represented. Degenerate and junk dimensions consolidate identifiers or flags, and role-playing dimensions reuse one dimension for several purposes.

Question 41Data mining

Which data mining technique groups records into clusters based on similarity without using predefined labels?

  1. Clustering Correct answer.
  2. Regression
  3. Classification
  4. Association rule mining

Explanation

Clustering is an unsupervised technique that partitions records so that members of the same group are more similar to each other than to records in other groups. It does not rely on a labeled target variable, which distinguishes it from supervised methods.

Question 42Data mining

A model is trained on historical loan records labeled as default or no default in order to predict the label for new applicants. Which task is this?

  1. Clustering
  2. Classification Correct answer.
  3. Anomaly detection
  4. Dimensionality reduction

Explanation

Classification assigns records to predefined categorical classes using a model learned from labeled examples. Because the historical records carry known default labels, the task is supervised classification.

Question 43Data mining

In market basket analysis, what does the support of an itemset represent?

  1. The share of antecedent transactions that also contain the consequent
  2. The count of distinct items in the transaction file
  3. The ratio of observed to expected co-occurrence
  4. The proportion of all transactions that contain the itemset Correct answer.

Explanation

Support measures how frequently an itemset appears across the entire transaction database, expressed as a proportion or count. It is the denominator basis used when computing confidence and lift.

Question 44Data mining

Which methodology defines six phases beginning with business understanding and ending with deployment?

  1. Agile
  2. CRISP-DM Correct answer.
  3. SEMMA
  4. KDD

Explanation

CRISP-DM is a six-phase process model: business understanding, data understanding, data preparation, modeling, evaluation, and deployment. Its lifecycle framing makes it the common reference for structuring analytics projects.

Question 45Data mining

Which task is an example of supervised learning?

  1. Grouping customers by purchase behavior without labels
  2. Predicting house price from labeled sales records Correct answer.
  3. Finding frequent item combinations in transactions
  4. Reducing variables with principal component analysis

Explanation

Supervised learning trains on records that include a known target value, then predicts that target for new records. Labeled sales records provide the outcome needed to learn a price prediction model.

Question 46Data mining

A confusion matrix is primarily used to do what?

  1. Estimate the number of clusters in a dataset
  2. Summarize correct and incorrect predictions of a classifier Correct answer.
  3. Measure the correlation between two numeric variables
  4. Rank association rules by support

Explanation

A confusion matrix cross-tabulates actual versus predicted classes, showing true positives, false positives, true negatives, and false negatives. Those counts feed metrics such as accuracy, precision, and recall.

Question 47Data mining

Which sampling method divides the population into homogeneous subgroups and samples from each in proportion to its size?

  1. Systematic sampling
  2. Cluster sampling
  3. Stratified sampling Correct answer.
  4. Simple random sampling

Explanation

Stratified sampling splits the population into strata that are internally similar and then draws from every stratum, often proportionally. This preserves the representation of each subgroup in the sample.

Question 48Data mining

Why is a dataset typically split into training and test subsets?

  1. To increase the number of records available for modeling
  2. To convert categorical variables into numeric form
  3. To estimate how well a model generalizes to unseen data Correct answer.
  4. To remove outliers before analysis

Explanation

Holding out a test subset lets the analyst measure performance on records the model did not learn from. That estimate indicates whether the model generalizes rather than memorizes the training data.

Question 49Data mining

When applying k-means clustering, what must the analyst specify before the algorithm runs?

  1. The labeled class of each record
  2. The correlation threshold between variables
  3. The number of iterations, fixed at 100
  4. The number of clusters, k Correct answer.

Explanation

K-means requires the target number of clusters k as an input, and the algorithm then iteratively assigns points and updates centroids. Analysts often compare solutions across several k values using measures such as the elbow method or silhouette score.

Question 50Data mining

An association rule A implies B has support(A and B) = 0.20 and support(A) = 0.50. What is the confidence of the rule?

  1. 0.70
  2. 0.10
  3. 0.50
  4. 0.40 Correct answer.

Explanation

Confidence equals the support of the combined itemset divided by the support of the antecedent: 0.20 divided by 0.50 equals 0.40. It estimates the share of A transactions that also contain B.