فیزیک | تمرین ۲

When several measurements lie close together, it is tempting to regard their agreement as evidence that the measured quantity has been determined correctly. Such agreement shows consistency under the conditions of measurement, but a shared displacement from a suitable reference can remain hidden in a tightly grouped set of results. Repeating a procedure many times need not expose an error that is reproduced by the procedure itself.

Imagine a thermometer that gives nearly identical readings whenever it is placed in the same stable environment, while comparison with a suitable reference reveals a consistent offset. The repeatability of its readings would not remove the need to investigate that offset, nor would averaging more of those readings necessarily bring the result closer to the reference. The example distinguishes variation among repeated readings from a systematic difference that those readings share.

A calibration record is useful because it documents a relationship established under specified conditions rather than merely certifying that an instrument has produced stable numbers. Whether that relationship remains suitable must be considered when the instrument is used in a substantially different environment or after its behavior has changed. Confidence in a reported measurement therefore depends on understanding both the spread of the observations and their relationship to an appropriate reference under the relevant conditions. Reporting those conditions helps another observer decide whether two results are comparable, especially when a disagreement might arise from a difference in the measurement arrangements rather than in the quantity itself. A useful report consequently preserves the grounds for comparison instead of offering a bare number as its entire justification.