Normal distribution and standardisation Flashcards
lower SD means
more consistent
if there is an error how do you report it
missing value
reasons to transform data
scale eg units
standardise eg by sex
what happens to the mean and SD when adding or subtracting constant numbers to each value in a data set
changes mean by amount added or subtracted
SD remains the same
when multiplying or dividing to a scale …
mean and SD increases or decreases by the proportion its being multiplied or divided by
Z score
number of SD an observation is from the mean
positive z score =
observation above the mean
negative z score =
observation lower than mean
zero z score =
observation equals mean
z score equation
observation - mean
/
SD
mean of z scores is always
zero
SD of z scores is always
1
area under a normal curve =
1
probability of observation being above mean =
0.5
68.2% of observation lie between …
-1 and +1 SD
95.4% of data falls between …
-2 and +2 SD
1 z score on a normal distribution is
34.1% or 0.341
normal table
cells show the proportion od the area under the curve that lies between the mean and a positive z score
first column of a normal table =
gives z score to 1 decimal place
how to travel a normal table
go down then along eg if a z score is 0.66, go down to 0.6 and then along to 0.06
what’s does the value you find from the normal table show
the number of people lie between mean and the z score you used