answersLogoWhite

0

Normalizing data If by "normalizing data" is meant the process by which data is transformed so that it more closely approximates a normal distribution, one method is to take the logarithm of the individual data points to the base 10. If by "normalizing data" is meant the process by which data is transformed so that it can be compared with other data from a different scale (standardization), one method is to convert the individual data points to Z scores. Z scores have a mean of zero. The individual data points are converted to numbers that are multiples or fractions of one standard deviation (SD). A datum that is equal to the mean gets a Z score of zero. A datum that is 1.5 SD above the mean gets a Z score of +1.5. A datum that is half a SD below the mean gets a Z sore of -0.5. Data Z score 60 -1.39 65 -1.04 70 -0.69 80 0.00 90 0.69 95 1.04 100 1.39 Mean: 80.0 SD: 14.4 The lefthand column is the raw data. The mean is 80, and the SD is 14.4. The Z scores -- the standardized data -- based on that mean and SD are in the righthand column. {| |}

User Avatar

Wiki User

17y ago

What else can I help you with?