Loading page content
Original practice bank · All explanations and revision guides free
Read 24 original SQL tasks with complete input tables, checked on SQLite 3.53.4. Free worked answers and four full guides. Untimed practice.
24 questions · about 12 minutes · no signup
Treat each task as a fresh SQLite 3.53.4 database. Run the displayed CREATE/INSERT setup in order, then evaluate only the statement after the comment. For SELECT answers, | separates columns and ; separates rows; no header is included, and NULL denotes a database NULL. Multirow queries specify their output order. For the two rejected INSERTs, select the constraint outcome rather than a row set. English identifiers and values stay identical across interfaces. There is no timer or speed bonus. All 24 valid choices are needed before personal counts appear. If you consult references, interpret the count under those conditions.
Preparing the test…
All questions and full results are free, with no signup. Saving is optional; only saved attempts can be resumed after a refresh or language change. Tab drafts expire after 24 hours and clear on completion or retake.
What rows does the final query return, or does the shown INSERT fail? Each of these 24 original SQL tasks supplies its own table definitions and input rows. A fresh SQLite 3.53.4 in-memory database was used for each check: 22 SELECT statements produced the keyed rows and two INSERT statements failed the stated constraints. Four groups cover filters and NULLs, joins and matches, groups and aggregates, and windows and constraints. Every explanation and all four full guides are free. You choose outcomes here; the site runs no visitor SQL and needs no account, upload or database credentials.
One matching key contributes one correct item; another valid choice contributes zero. Each group has six items, making a total out of 24. All choices must be valid before personal counts appear. There is no time weighting, partial credit or silent completion of missing answers. A 4/6 section count means four matching outcomes on those supplied tasks, without a percentile, hiring pass mark or qualification.
Every section has a public revision guide. Completing the practice shows every guide and item explanation without assigning a type by score. Public guides contain no personal responses or counts; studying before practice is welcome. This is not a secure exam.
Review numeric and zone filters, IS NULL, equality with NULL, inclusive BETWEEN and a COALESCE fallback.
Review inner and left joins, ON versus WHERE, NULL-padded rows, card counts and correlated NOT EXISTS.
Review row and value counts, grouped sums, HAVING, filtered totals, distinct counts and an outer maximum.
Review ROW_NUMBER, ranking ties, running totals, a filtered CTE and two explicit INSERT constraint failures.
Each task was checked with SQLite 3.53.4 in a fresh in-memory database, using only its displayed schema and invented rows. The final statement follows the setup. This is a declared environment, not a claim that all SQL engines and settings behave identically.
A vertical bar separates columns and a semicolon separates rows. No heading row is counted. NULL denotes a database NULL. Each multirow query states its order. The two constraint questions ask whether the final INSERT is rejected rather than showing SELECT rows.
No. Read the supplied tables and statements, then select an outcome. There is no visitor SQL execution, private database connection or file upload. English code and values remain the same across interfaces, with bilingual instructions and explanations.
Yes. A complete valid attempt shows every selected option, key and worked explanation at any count. All four public guides are also readable before starting. No email, signup, payment or certificate purchase is required.
It counts matching outcomes in this fixed original bank, without professional bands, norms, employer cutoffs or qualifications. There is no timer or speed bonus. Around twelve minutes is an editorial estimate. Reading keys before retaking can improve memory without demonstrating transfer.
Answers stay in the browser. Optional tab drafts expire after 24 hours and clear on completion, exit-and-clear or retake; refresh loses progress without one. Sharing copies only a public practice link, without answers, counts, query submissions or private identifiers.
The result includes every original setup, final statement, selected option, key and explanation. All four full guides are available at zero, partial and perfect counts. Column and row separators are a display convention, not extra returned values. Review the actual input rows and query clauses when tracing a mismatch. Public guides contain complete revision content rather than a personal score.
Every task was checked in its own fresh in-memory SQLite 3.53.4 database with invented input, not a production database. SQL dialects, settings, constraints and concurrency behaviour can differ; this bank does not establish identical behaviour across engines or assess query tuning, administration or live reporting. It has no timer. Around twelve minutes is an editorial estimate, and retake repeats the same fixed bank.
FermatMind wrote the table data, questions, distractors, worked explanations and guides. The linked documentation supports selected SQL rules; its examples and proprietary recruitment banks are not reproduced. This independent practice has no vendor endorsement or certification. Code is displayed as escaped text, and choices are checked locally. No visitor query is executed or sent to a connected database.
These sources provide background, not validation of this questionnaire or an individual diagnosis.
Website content edition 1.0.0 ·
Source question / scoring / report versions: fermatmind_sqlite353_original_24_v1 / practice_equal_correct_counts_v1 / fm_sql_four_full_revision_guides_v1