11.520: A Workshop On Geographic Information Systems 11.188: Urban Planning and Social Science Laboratory Comments On Lab 1 and Lab2
11.520: A Workshop On Geographic Information Systems 11.188: Urban Planning and Social Science Laboratory Comments On Lab 1 and Lab2
From the quality of handed homework of Lab 1 and Lab 2, we can see that our class is off
to a very good start. Most people have learned the basic skills to get started with ArcGIS
and got good grades in the first two labs. We put some comments on most of the
homeworks. Here, we will summarize the common problems in the handed labs and give
some examples in two grading levels: 'check+' and 'check.'
Common problems:
1.Readablity of the map
We use maps to convey information to readers in an efficient, visual manner. To help the
readers understand our intentions, we need an easily readable map that focuses attention
on the main storyline and reasoning. To accomplish this, you need a clear map title,
organized map legend, succinct map data-source/date/scale/north-arrow, and readable
maps that highlight a defensible characterization of the spatial patterns and relationships.
These are basic steps in map making but for want of time or some other reason, some of
the work shortchanged some of these concerns and had overlapping map and legend;
inappropriate symbol scales that make the hard to read; no title, etc. Also, think about
how the real estate of the page is utilized to convey your 'story'. For example, are the
important map gradations concentrated within a small thematic map in one corner of the
page instead of being rearranged to allow a larger map? Are the colors and symbols
chosen so that your eye is drawn to the important elements or contrasts in the map
(without exaggerating the differences by using some of the bad techniques mentioned in
'How to Lie With Maps')?
2. Precision of the map
It is very easy to make some mistakes during map making, and it is relatively hard to
check these errors when maps are done. In these labs, you are required to follow the
instructions of the labs. But we can see that some people missed some important part: the
precision of map - for example, listing real price in the legend, but showing something
else on the map; not following the instruction and using something other than quantiles to
classify the thematic map. It is not a big deal in the first labs, but we hope the class will
double-check whether they are accurately identifying what was done and following the
lab instructions. Using standard deviation instead of quantile - and making that clear - is
defensible, but miss-labeling the map or using the default classification (natural break)
without making that clear is a more serious error. In general, it is a good idea to annotate
the legend so that the classification method is clarified (just putting 'quantile' in
parentheses under the legend title is often enough, and quite informative).
3. Art of the map
It is easy to undertand that for a map the more beautiful, the better. To be sure, there is an
art as well as a science to good map making, and all 'experts' don't agree on one 'right' or
'best' way to present the data. Nevertheless, after you make sure that you have the right
contents and necessary elements, spend a little time attending to the graphic design of
your maps, and you will have an improved map that does a much better job of conveying
the desired information to the reader.