Transcript PowerPoint - Crescent School

Mapping Classification
techniques
Groups of Features with Similar
Values
Geographer's WorkBench G.E.M. Geotechnologies 2001
Natural Breaks
• Finds groupings and patterns inherent
in your data.
• Values in a class are likely to be similar
• Data values that cluster are placed into
a single class
• Class breaks are defined where there is
a gap between clusters of values
Geographer's WorkBench G.E.M. Geotechnologies 2001
Natural Breaks
How it Works;
• GIS determines high
and low value for each
class
• Uses mathematical
procedure to test class
breaks
• Picks class breaks to
best group similar
values & maximize
differences between
classes
What it is Good for:
• Mapping data values
that are not evenly
distributed
Disadvantages
• Data sets are map
specific,therefore hard
to compare to other
maps
• Choosing optimum
number of classes is
difficult
Geographer's WorkBench G.E.M. Geotechnologies 2001
Natural
Breaks
Canada Population
Natural Breaks
26745 - 137312
137313 - 729498
729499 - 1119583
1119584 - 3907738
3907739 - 11410046
Notice how the population has been separated in this Natural Breaks layout.
The division takes place where there is an obvious difference in population.
Geographer's WorkBench G.E.M. Geotechnologies 2001
Quantile
• Each Class has an equal number of
features in it.
How it Works
• GIS orders the features, based on attribute
value,usually from low to high.
• Sums the number of features
• Divides by the total number of classes you have
pre chosen
• Fills all the classes with an equal number of
features
Geographer's WorkBench G.E.M. Geotechnologies 2001
Quantile
What is it good for:
• Comparing areas that
are roughly the same
size
• Mapping data in which
the values are evenly
distributed
• Emphasizing relative
position of a feature
among other features IE
top 20 %
Disadvantages:
• Features with close
values may end up in
different classes
• Can exaggerate the
difference between
features
• Conversely few widely
ranging value may end
up in same class
Geographer's WorkBench G.E.M. Geotechnologies 2001
Quantile
Canada Population
Quantile
26745 - 37360
37361 - 729498
729499 - 1119583
1119584 - 3907738
3907739 - 11410046
This is very similar to Natural Breaks, however the categories are quite different.
The groupings are more equal in their amount of population than Natural Breaks.
Geographer's WorkBench G.E.M. Geotechnologies 2001
Equal Interval
• Each class has an equal range of values –
that is the difference between the high and
low value is the same for each class
How it works:
GIS subtracts lowest value in data set from highest
Divides that number by the number of classes you have chosen
adds that number to the lowest data value to get maximum value
for the first class
Then adds to each maximum value to set breaks for rest of the
classes
Geographer's WorkBench G.E.M. Geotechnologies 2001
Equal Interval
What it is good for:
Disadvantages:
• Presenting information to a
non-technical audience
• Equal intervals are easier
to interpret since the range
for each class is equal
• Good with familiar values%
• Good for mapping
continuous data –
precipitation, temperature
• Problem is data is
clustered rather than
evenly distributed
• There may be too many
features in one or two
classes and some
classes with no features
Geographer's WorkBench G.E.M. Geotechnologies 2001
Equal
Interval
Canada Population
Equal Interval
26745 - 2303405
2303406 - 4580065
4580066 - 6856726
6856727 - 9133386
9133387 - 11410046
With Equal Interval the population has been divided so there is equal
numbers for each category. This is quite different from the previous two.
Geographer's WorkBench G.E.M. Geotechnologies 2001
Standard Deviation
Each class is defined by its distance
from the mean value of all the feature
How it works:
• Through a formula it measures the average
amount that data varies from the mean
• GIS creates class breaks above and below the
mean based on the number of standard
deviations you specify such as ½ or 1 standard
deviation.
Geographer's WorkBench G.E.M. Geotechnologies 2001
Standard Deviation
What it’s good for:
Disadvantages:
• Seeing which features
are above and below
an average value
• Displaying data that
has many values
around a mean and
few further from the
mean – Bell Curve or
normal distribution
• The map doesn’t
show the the actual
values of the features,
only how far their
value is from the
mean
• Very high or low
values can skew the
mean so that most
features will fall in the
same class
Geographer's WorkBench G.E.M. Geotechnologies 2001
Standard
Deviation
Canada Population
Standard Deviation
< -1 Std. Dev.
-1 - 1 Std. Dev.
1 - 2 Std. Dev.
> 2 Std. Dev.
With Standard Deviation the population has been plotted by how far away
from the mean each province is. Here we can see what provinces are
above or below the mean easily.
Geographer's WorkBench G.E.M. Geotechnologies 2001
Choosing a Classification
Scheme
You need to know how data values are
distributed
Methodology:
• Create bar graph
• Set horizontal axis to be attribute levels
• Vertical axis should represent # of features
having a particular value
• Choose proper classification scheme
Geographer's WorkBench G.E.M. Geotechnologies 2001
Choosing a Classification
Scheme
• If your Data is unevenly distributed
• many features have the same or similar
values,
• there are gaps between groups of values
NATURAL BREAKS
Geographer's WorkBench G.E.M. Geotechnologies 2001
Choosing a Classification
Scheme
• If your data is evenly distributed
• You want to emphasize the relative
difference between features
QUANTILE
Geographer's WorkBench G.E.M. Geotechnologies 2001
Choosing a Classification
Scheme
• If your data is evenly distributed
• You want to emphasize the difference
between features
EQUAL INTERVAL
Or
STANDARD DEVIATION
Geographer's WorkBench G.E.M. Geotechnologies 2001