This is untimed educational practice for one applied skill. It is not a clinical, diagnostic, employment, or professionally recognized assessment, it does not produce a score or credential, and for the judgement and safety families it is not a substitute for your employer's own policies, training, or legal duties.
Each item is about accuracy, technique, transcription rules, or how typing tests are scored: the parts of keyboard skill that reading and reasoning can improve. Speed itself needs a keyboard and a timer, not a quiz. Nothing here is timed.
Choose one answer for each item. On a judgement item, more than one option is often defensible. Pick the one you would rate most effective, then check the worked explanation for why the runner-up is weaker.
Warm up on the character-level comparison that decides an accuracy score. Read the source, then read the transcript against it rather than for sense.
1. typing-keyboard-applied-s01-q01
Source line: Invoice 4820-B, due 14/03, GBP 1,275.40. Which transcript reproduces it exactly?
Answer:
2. typing-keyboard-applied-s01-q02
Source: 'Please confirm receipt by 5pm on Thursday.' Transcript: 'Please confirm reciept by 5pm on Thursday.' What kind of error is this?
Answer:
3. typing-keyboard-applied-s01-q03
Source: 'the committee's recommendation'. Transcript: 'the committees recommendation'. What has changed?
Answer:
4. typing-keyboard-applied-s01-q04
What is the most reliable way to catch a transposed digit inside a long reference number?
Answer:
5. typing-keyboard-applied-s01-q05
Autocorrect has silently changed 'form' to 'from' in a document you typed. What is the lesson?
Answer:
The physical layer. These are the habits that decide whether speed is sustainable or merely achievable for ten minutes.
6. typing-keyboard-applied-s02-q01
On a QWERTY keyboard, where does the left index finger rest in the touch-typing home position?
Answer:
7. typing-keyboard-applied-s02-q02
Why do the F and J keys carry small raised bars?
Answer:
8. typing-keyboard-applied-s02-q03
Which set of habits best supports comfortable typing over a full working day?
Answer:
9. typing-keyboard-applied-s02-q04
Why is a numeric keypad usually faster than the number row for long columns of figures?
Answer:
10. typing-keyboard-applied-s02-q05
What is the strongest reason to learn keyboard shortcuts for the ten operations you repeat most often?
Answer:
Harder items about fidelity. A transcriber's job is to reproduce the source and flag what is wrong with it, not to improve it.
11. typing-keyboard-applied-s03-q01
On a handwritten form, one digit could plausibly be a 1 or a 7. The field is an account number.
Answer:
12. typing-keyboard-applied-s03-q02
A date field requires the format DD/MM/YYYY. The source document reads '3 Feb 24'. What do you enter?
Answer:
13. typing-keyboard-applied-s03-q03
While transcribing, you find that the source document contains an obviously invalid postcode.
Answer:
14. typing-keyboard-applied-s03-q04
When does double-keying, two people entering the same data independently, earn its cost?
Answer:
15. typing-keyboard-applied-s03-q05
Five hundred records must be entered by the end of the day. You are behind, and the verification step is what would have to go.
Answer:
Finish with the measurement itself: what a words-per-minute figure counts, and what actually makes it improve.
16. typing-keyboard-applied-s04-q01
A typing test reports 60 words per minute gross and 52 net. What is the difference?
Answer:
17. typing-keyboard-applied-s04-q02
In words-per-minute scoring, how is a 'word' defined?
Answer:
18. typing-keyboard-applied-s04-q03
You can type at 70 wpm with about four errors a minute, or at 55 wpm with essentially none. Which is more valuable for data-entry work?
Answer:
19. typing-keyboard-applied-s04-q04
What is the most effective way to raise typing speed over a few weeks?
Answer:
20. typing-keyboard-applied-s04-q05
Your speed drops sharply when the passage contains unfamiliar technical terms. What does that tell you?
Answer:
This applied practice source does not publish separate authored hints: the rationale is attached to each item as a worked explanation instead. Every item is therefore listed as missing a hint, and no hint text is generated or inferred.
20 worksheet items do not have an authored hint.
This source contains no authored hints. No hint text was generated.
| Item | Choice | Answer |
|---|---|---|
| typing-keyboard-applied-s01-q01 | D | Invoice 4820-B, due 14/03, GBP 1,275.40 |
| typing-keyboard-applied-s01-q02 | B | A transposition: two adjacent letters typed in the wrong order |
| typing-keyboard-applied-s01-q03 | C | A dropped apostrophe has turned a possessive singular into a plural |
| typing-keyboard-applied-s01-q04 | A | Compare it against the source in small groups of digits, or use a field with a check digit that rejects an invalid number |
| typing-keyboard-applied-s01-q05 | B | Proofread the finished text yourself: a correctly spelled word in the wrong place is invisible to a spell-checker |
| typing-keyboard-applied-s02-q01 | C | On F |
| typing-keyboard-applied-s02-q02 | A | They let a touch typist find the home position by feel, without looking down |
| typing-keyboard-applied-s02-q03 | D | Keeping wrists straight rather than bent, forearms roughly level, the screen near eye height, and taking short breaks regularly |
| typing-keyboard-applied-s02-q04 | B | All ten digits and the common operators sit under one hand with almost no movement, and the 5 key carries a locating bump for touch entry |
| typing-keyboard-applied-s02-q05 | D | Each use removes a hand-to-mouse round trip and a visual search, and that is where most of the time in repetitive work goes |
| typing-keyboard-applied-s03-q01 | A | Flag it for verification and record the uncertainty rather than choose between them |
| typing-keyboard-applied-s03-q02 | D | 03/02/2024 |
| typing-keyboard-applied-s03-q03 | B | Transcribe it exactly as given and raise a query, because the job is fidelity to the source plus a flag |
| typing-keyboard-applied-s03-q04 | C | When an undetected error would be costly and no automatic validation can catch it |
| typing-keyboard-applied-s03-q05 | A | Keep the checking discipline and report that the batch will be short, because an error costs more downstream than the delay does |
| typing-keyboard-applied-s04-q01 | B | Net applies a penalty for errors, so it is the figure that reflects usable output |
| typing-keyboard-applied-s04-q02 | C | A standard unit of five keystrokes, including spaces |
| typing-keyboard-applied-s04-q03 | A | 55 with no errors, because finding and correcting an error costs far more than the extra speed saved |
| typing-keyboard-applied-s04-q04 | D | Practise at a pace you can hold without looking down and without errors, drilling the specific sequences you get wrong; speed follows accuracy |
| typing-keyboard-applied-s04-q05 | C | It is expected: typing speed is largely learned sequences, so unfamiliar strings have to be practised specifically |
The others each contain exactly one error: transposed final digits, the letter B typed as the digit 8, and the decimal and thousands separators swapped. All three read plausibly at a glance, which is why an accuracy check has to compare characters rather than impressions.
The 'ei' has been typed as 'ie': the two characters are present but reversed, which is the single most common typing error and the hardest to see when reading for meaning. It is worth classifying as a transposition rather than a spelling mistake, because the drill that fixes it is different: slow, accurate repetition of the specific sequence.
One committee's recommendation and several committees' recommendation are different claims, so the missing character changes the meaning rather than the appearance. Treating the apostrophe as optional is the tempting judgement in fast typing, and it is exactly the class of error a transcription check is looking for.
Short groups fit in working memory, and a check digit catches the error automatically at the moment of entry. Re-typing from memory is the strongest wrong answer because double entry is a genuine control, but it only works when both entries are taken from the source; from memory you simply reproduce the same misreading and gain false confidence.
Spell-checkers compare words against a dictionary, and both 'form' and 'from' are in it, so no amount of attention to underlined words will surface this class of error. Turning the checker off is the overcorrection: it removes a tool that catches a different and larger set of mistakes without helping with this one at all.
The left hand rests on A, S, D, F, with the index finger on F; G is reached by stretching that same finger one key right. Choosing G is the natural error because the index finger does spend time there, but a home key is where the finger returns, not the furthest key it covers.
Two tactile landmarks are enough to place eight fingers, which is what makes returning to the home row possible without breaking eye contact with the source. The frequency answer is plausible and wrong on the facts: E, T, and A are far commoner than F and J, which is precisely why those two could be spared as landmarks.
Neutral joints, a screen that does not force the neck to bend, and frequent short breaks address posture and static loading together. Resting the wrists while typing is the near-miss and is actively unhelpful: it anchors the hands so the fingers must stretch and the wrists bend, and it presses soft tissue against a hard edge. A wrist rest is for pauses, not for keystrokes.
A compact cluster plus a tactile centre key means the hand never leaves position and the eyes never leave the source. The registration-speed answer sounds technical and is simply untrue. The keys are the same switches wired to the same controller.
The saving is not the keystroke, it is the hand leaving the keyboard, the eyes hunting for a target, and the hand coming back: several seconds, repeated hundreds of times a day. The 'works everywhere' answer is the tempting one and is false: shortcuts vary between applications and platforms, which is why the advice is to learn your ten, not to learn them all.
A flagged ambiguity gets resolved by someone who can check the account; a guess enters the system indistinguishable from certain data. Leaving it blank is the honest-looking alternative and loses information: an empty field says nothing about which two values were possible or that anyone looked.
Day and month are padded to two digits and the year is written in full, which is what the specified format requires. The important trap is 02/03/2024: that is the same date written US-style with the month first, and a value that reads as valid under both conventions is one nobody downstream will ever query.
Transcription and correction are separate responsibilities, and mixing them makes it impossible afterwards to tell what the original document actually said. Correcting it to the intended value is the helpful-seeming option and is how a transcription silently becomes an interpretation that nobody can audit.
The test is the cost of the error against the cost of the second pass, and validation rules or check digits are usually cheaper when they are available at all. 'Always' is the accuracy-first instinct and ignores that double-keying every field would consume the budget that could have bought validation for the fields that matter.
A short batch is a visible, correctable shortfall; an unverified batch is an invisible one that surfaces weeks later inside somebody else's work. Dropping verification is the option that makes the day's target and moves the cost to a point where nobody can trace it back to this decision.
Gross is raw throughput; net deducts for mistakes, which is why employers quote net and why a large gap matters more than a modest gross figure. The 'different passage' answer is plausible if you have never seen the two reported together, and both numbers always come from the same run.
The five-keystroke standard is what makes scores comparable between passages of different vocabulary. Counting actual words between spaces is the intuitive definition and would make a passage of short words score far higher than one of long ones, which is exactly the distortion the convention removes.
Four errors a minute is a couple of hundred a day, and each one costs a detection, a correction, and sometimes a downstream consequence far larger than either. The threshold answer is the sophisticated-sounding one and mistakes the scoreboard for the game: thresholds exist, but they are set precisely because errors are expensive.
Typing is a motor skill built from learned sequences, and practising errors trains them in just as reliably as practising correct movements. Pushing for maximum speed is the intuitive training method and is the one that produces a plateau: the errors become automatic alongside the speed, and unlearning them takes longer than learning them did.
Fluent typing runs on chunks, common letter groups executed as single movements, and an unfamiliar term has no chunk, so it falls back to letter-by-letter. Concluding your general figure was overstated is the natural inference and is wrong: both measurements are accurate, they are just measuring different material.