Dispersion Flashcards
The average is a good starting point, but we usually need more than one number to help us make sense of data.
Critically we need to know how much the data varies, or how:
dispersed it is
Dispersion is concerned with how spread out or:
‘dispersed’ values are
Can measure dispersion in different ways, with different ways of doing this linked to different measures of the:
average
If we are using a mean to measure the average we can also use a
STANDARD DEVIATION
If we are using a median to find the mid point we can also use
PERCENTILES
Average (or ‘central tendency’) can be the same for each distribution in groups/data sets, but VARIATION or DISPERSION can be:
different
Homogenous occurs when:
cases in data are of the same value, e.g. all 6
Dispersed is when the cases in the data are
spread out/dispersed over a wide area
Clustered are when the cases in the data are
positioned close together
STANDARD DEVIATION is a measure of how far away, on average, data points are from the
mean
e.g. if mean is 6 and two points are on 4 then they are -2 away, three points on 7 they are +1 away
Work out standard deviation:
- multiply the number of cases by distance away squared,
e. g. 2*-2squared - add up squared and multiplied numbers
- divide by number of cases
- square root total = SD
Mean is a statistical way of working out the average, and so is the:
SD
In contrast, the median refers to position =
the middle value
We can also talk about dispersion by talking about:
position
Can split histogram into 10 parts. Each part is a
decile
