|
average january temperatures across the US |
 |
|
this data
>> was taken from a sample of 56 cities in the US. every year for 30
years, the average daytime temperature for the month of january was recorded in each city. the
y-axis indicates how many cities fall into the 'categories' on the
x-axis.
individual values ranged from 0° all the way up to 65° -
with the mean at about 26.49°F.
50% of the data lies between
18° and 34° and the two highest values (58° and
65°) can be considered outliers. |
|
|
the data can be
considered a normal
distribution: when
plotted as a histogram, it roughly follows a bell curve- with a distinct peak towards the
center and a symmetric
shape. |
the bell
curve:
 |
|
|
<< a smooth
curve can be used to illustrate the overall shape of the
distribution.
the majority of the
data is within 1-2 standard deviations from the mean. |
* the standard deviation from the mean in this data set is
approximately 13.50°F.
* the standardized score
(z-score) of
seattle is 1.34 (18.09°F above the
mean). |
|
|
at
44°F, seattle is 18.09°F above the mean
temperature. this puts us at the 90.1 percentile mark. this is particularly
interesting because while we consider seattle to be a pretty cold place in
january, we usually don't realize that we are warmer than about 90% of the
rest of the of the country.
|
|
|
when scores are
transformed from a normal distribution into standardized scores
(z-scores), they are changed from being proportionate to a specific number
to being proportionate to a mean of zero. the shape of the
distribution remains the same. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|