Skip to main content
unnormalized

unnormalized

1 7.9
Not adjusted to the same scale as other values
  • adjective
  • /ʌnˈnɔrməˌlaɪzd/
  • Specialized
: no normalizado
  • In some mathematical models, the values are often presented in an unnormalized form to simplify calculations.

Examples

  • This main effect was not significant in the initial analysis of the unnormalized RT data.

    Academic text (1997)
  • Zero, infinity, NANs, and unnormalized and denormalized numbers cannot use the data bypasses.

    Academic text (1994)
  • You might actually use it in an unnormalized fashion.

  • You do IG of A, it'll give you the eigenvalues of matrix A and it'll give them back to you unnormalized and unordered.

  • The data set contains unnormalized scores, which require adjustment for proper analysis.

  • To compare the results more effectively, we need to convert the unnormalized figures into a standardized scale.

Synonyms

uncalibrated
vsunnormalized
  • Specialized
1 6.6

Not adjusted to give correct measurements so results may be wrong

means not scaled to a standard value rather than just instrument tuning
unscaled
vsunnormalized
1 6.0

Not measured, changed, or covered with scales, for example raw data and fish

focuses on missing scale adjustment rather than instrument accuracy
unstandardized
vsunnormalized
  • Specialized
1 6.8

Expressed in original units rather than a common scale

is about not being adjusted to standard values across math and computing
nonstandardized
vsunnormalized
  • Specialized
8.1

Values shown in their original measurement units and not changed to a common scale

is similar but also applies beyond measurements to computing and statistics

How Normalized

standardized
  • US
61 4.0
unnormalized
  • Specialized
1 7.9
raw
455 1.0

Surface Forms

unnormalized positive

Morphology

unnormalized = normalized (transparent) = un + normal + ize + ed

Formed by adding the negating prefix 'un-' to 'normalized'; the derivation follows regular, productive morphology.

Etymology

Unnormalized is built from the prefix un- meaning 'not' and normal meaning 'standard' or 'usual'. So an unnormalized number or set of data is simply 'not made normal' — it has not been adjusted or scaled to the same standard, which is why you often change it before you compare or use it.