Browse the glossary using this index
Special |
A |
B |
C |
D |
E |
F |
G |
H |
I |
J |
K |
L |
M |
N |
O |
P |
Q |
R |
S |
T |
U |
V |
W |
X |
Y |
Z |
ALL
Quantiles A method of data classification that attempts to place an equal number of values into each class. For example, if there were 50 observations, each with a different value so that no duplicates were encountered during data classification, grouping them into five classes (also known as quintiles) would result in 10 observations per class. Quantiles are useful for classifying ordinal (rank-order) data, and for comparing maps with the same number of classes. Outlier values will be less visible and attention will be focused on relative rankings. The map resulting from this method of data classification will tend to produce an even distribution of map colors. Caution: If the data is highly skewed, the quantiles classification method will still place the data into the number of classes specified. This forcing of unevenly distributed data into classes containing equal numbers of observations could lead to the false impression that the data is normally distributed.
|
|