3. Recording Error
3.1. Examples
Reading the wrong scale on a dual-scale instrument (e.g. Fahrenheit instead of Celsius)
Misreading a burette as 23.4 mL instead of 32.4 mL (digit transposition)
Writing down the wrong units (e.g. recording grams when the balance reads milligrams)
Copying a result incorrectly from the bench into a data table
Recording a result from the wrong trial in the wrong row of a table
Recording Errors: A Four-Step Analysis
Use the four-step framework to analyse any recording error:
- Step 1 — Identify the source
Observation / procedure — the researcher made a mistake when reading an instrument or transcribing a value, rather than in the experimental technique itself.
- Step 2 — Classify the behaviour
A one-off mistake → Personal error → produces an invalid result; discard and repeat.
- Step 3 — Explain the impact
The recorded value does not represent the true measurement. It is an invalid result and should not be included when calculating a mean or assessing uncertainty.
- Step 4 — Suggest an improvement
Recording errors are eliminated, not reduced — check the record immediately and re-read or repeat the measurement correctly.
3.2. Effects
Recording errors produce invalid results. Because they are caused by a one-off mistake rather than a consistent or random influence, they:
do not affect the precision of the remaining valid results,
do not introduce systematic bias,
should not be included when calculating a mean or assessing uncertainty, as the data were not recorded correctly.
3.3. Improvements
To eliminate recording errors, check each data point at the time it is recorded rather than after the experiment is complete.
Discard the invalid result and re-read or repeat the measurement correctly.
Check each value immediately after recording — confirm units, decimal place, and scale before moving to the next trial.
Have a second person check critical readings or transcriptions at the time they occur.
Use a structured data table prepared before the experiment so results are recorded in the correct row and column.
Read instruments at eye level and state the reading aloud before writing it down, to catch misreads before they are committed to the table.
Structured Question: Recording Error
A Year 8 class is investigating whether the height from which a ball is dropped affects the time it takes to hit the ground. Each group drops a ball from four different heights — 0.5 m, 1.0 m, 1.5 m, and 2.0 m — and records the fall time using a stopwatch. After completing all trials, one student copies the results from their rough notes into the class data table. When transferring the result for the 1.5 m trial, the student writes 0.45 s instead of the correct value of 0.54 s, transposing the last two digits.
(a) Identify the type of error made by the student and classify it as random, systematic, or personal. (2 marks)
(b) Explain how this error would affect the result for the 1.5 m trial. In your answer, refer to the direction of the error and its effect on the validity of that result. (3 marks)
(c) The group repeats all trials three times and calculates a mean fall time for each height. Evaluate whether this would reduce the effect of the error identified in part (a). (2 marks)
(d) Describe one improvement the student could make to prevent this error from occurring. (1 mark)
Reveal Answer Key
(a)
The error is a recording error, classified as a personal error.
(1 mark for naming recording error; 1 mark for personal error)
(b)
Because the digits were transposed when copying, the recorded value of 0.45 s is lower than the true value of 0.54 s — the fall time for the 1.5 m trial is underestimated by 0.09 s. (1 mark)
This result is invalid — it does not represent the true fall time because the data was not recorded correctly. The measurement itself was taken correctly; the error occurred only when the result was transferred to the data table. (1 mark)
Unlike a systematic error, this mistake affects only the single value that was copied incorrectly. All other results in the data table are unaffected. (1 mark)
(c)
Repeating the trials and averaging would not reliably reduce the effect of this error. (1 mark)
The recorded value of 0.45 s is invalid and should not be included in the mean. If it is included, it will pull the mean fall time for the 1.5 m condition lower than the true value. Averaging valid results with an invalid result does not correct the error — it distorts the mean. The correct approach is to identify the incorrect entry, return to the original rough notes, and correct the recorded value. (1 mark)
(d)
The student should check each value immediately after transferring it — reading the copied entry back against the original rough notes before moving to the next result — to catch any transcription mistakes at the time they occur. (1 mark)
Part (a): accept “recording error” or “transcription error.” Do not accept “operator error” — the measurement was taken correctly; the mistake occurred only when the value was copied into the data table. Do not accept “random error” — the mistake occurred in a single specific entry, not as an unpredictable variation across all trials. Do not accept “systematic error” — the error does not affect every measurement in the same direction.
Part (b): as with the operator error question, this question asks about validity rather than accuracy and precision, which is the correct framing for a personal error affecting a single result. Award the direction mark for correctly identifying underestimation and linking it to the transposition (0.45 s instead of 0.54 s). Award the validity mark for correctly stating the result is invalid because it was not recorded correctly. Award the third mark for correctly noting that only the single copied value is affected — the original measurement and all other entries are unaffected.
Part (c): as with the operator error question, averaging does not help because the affected result is invalid, not just imprecise. Award both marks only if the student correctly identifies that the invalid result should be corrected from the original notes or excluded, and explains why including it distorts the mean. A response that states the original rough notes can be consulted to recover the correct value should be awarded full marks.
Part (d): accept “have a second student check each transferred value against the original notes” or “record results directly into the final data table rather than copying from rough notes” as alternative valid responses. Do not accept “repeat measurements” — the measurement was taken correctly and repeating it does not address the transcription mistake.