The limit on the precision of a measurement. Analytical balances, for example, can weigh an object with an uncertainty of ñ0.001 or ñ0.0001 grams.

Uncertainty results from a lack of perfect knowledge of many factors that affect stock assessments, estimation of reference points, and management. Rosenberg and Restrepo (1994) identify 5 types: measurement error (in observed quantities), process error (or natural population variability), model error (mis-specification of assumed values or model structure), estimation error (in population parameters or reference points, due to any of the preceding types of errors), and implementation error (or the inability to achieve targets exactly for whatever reason).

In a nuclear decay measurement, uncertainty refers to the lack of complete knowledge of a sample's decay rate due to the random nature of the decay process and the finite length of time used to count the sample.

parameter associated with the results of a measurement, that characterizes the dispersion of the values that could reasonably be attributed to the measurand.

A measure of the amount of doubt or distrust with which the data should be used.

A measure of variety. Uncertainty is zero when all elements are in the same category. Uncertainty increases with both the number of categories and their equiprobability. [Principia Cybernetica Web

a range, estimated by the experimenter, that is likely to contain the true value of whatever is being measured

error; an estimate of the differences in values from test to test that are divided into two types, systematic and random, depending on their origin.

(or absolute uncertainty): suppose that a physical property is measured and that the value of the measurement is x. In the everyday use of the word, there is clearly uncertainty as to whether x is the "true" dimension measured, or whether the true value is different because of deficiencies in the means used to make the measurement (eg due to the lack of precision of the measuring instrument used or variations in the temperature at which the readings were taken). However, if many measurements are taken of what is being measured, it is reasonable to say that there is a "best" value (x best) equal to the average of all the measurements taken. In considering the differences between the various values of x taken and the average (x best), we may conclude that there is an uncertainty, known as dx, between any individual measurement of x and (x best). Note that the fractional uncertainty (or relative uncertainty, or precision) equals dx / (x best). Note also that the percentage uncertainty = the fractional uncertainty x 100%. For an alternative explanation of the foregoing, see Accuracy.

The degree of accuracy of the measuring method. For radioactive measurements, the uncertainty is a summation of the uncertainty in the measurement of the sample, measurement of the background, and other possible sources of error.

A parameter associated with the result of a measurement that characterises the dispersion of values that could reasonably be attributed to the measurand.

the characteristic reflecting the fact that any measurement involves estimates and cannot be exactly reproduced.

The degree to which a parameter or characteristic cannot be specifically determined.

The standard deviation of a sufficiently large number of measurements of the same quantity by the same instrument or method. The non-correctable part of the inaccuracy of an instrument, it represents the limit of measurement precision. The uncertainty of an instrument is caused by the unpredictable effects upon its performance of such factors as friction, backlash, and electronic noise.

The estimated bounds of the deviation from the mean value, expressed generally as a percentage of the mean value. Taken ordinarily as the sum of (1) the random errors (errors of precision) at the 95% confidence level, and (2) the estimated upper bound of the systematic error (errors of accuracy).

A measure used to quantify the plausible maximum and minimum values for emissions from any source, given the biases inherent in the methods used to calculate a point estimate and known sources of error.

An estimate of the range of values about the measured value in which the accepted value is believed to lie. A total uncertainty value represents the combination of a measure of the random error and estimated bounds to the systematic error. Considers all factors contributory to the measurement uncertainty. The standard deviation is the statistical tool used to calculate uncertainty.