02.03.10 Deliverable #2

+

I struggled with this answer:

Which classification do you think best represents the data and why?

Let's just say it was a process of elimination and list the reasons why I did not choose a classification:

1. I thought the standard deviation data was best until I realized the data curve was not bell-shaped and data was skewed rather than normally distributed.
2. Then I looked at the equal interval classification. While the data here was grouped into very familiar and easy to understand classes (25%, 50% etc...) the method, like quantile classification, does not consider the distribution of the data along the number line.

The Natural Break method is the best classification method to represent the percentage or proportion data for the following reasons: not only is it easy to understand and considers the distribution of data along the number line but it also places contiguous enumeration units into the same class, maximes spatial auto correlation and equalizes the area.

02.03.10 Deliverable #1

With the exception of perfecting the layout and determining which map is best for the data presented, this exercise was relatively easy.

I think there are a couple of reasons for this:

1. I read all the material for both this class and GIS before attempting the lab. I think I was able to understand some of the concepts better this way.
2. I completed the GIS lab first and I hit this topic as I was working my way though that lab exercise.

Problems:

1. When I inserted the legend for standard deviation the color ramp defaulted to one I was not using and the color ramp I was using for the other maps was unavailable. I manually changed each color to match the color ramp used in the other maps.
2. I could not seem to standardize each legend window to be the same window and font size. Any tips here would be greatly appreciated!
3. Determining the best map for the data presented -- statistics was a tough class for me and I am struggling with the answer to this question.
4. It is frustrating when each time you resize a data frame, the scale readjusts and I need to change it back to 1:825,000. Is there a way to prevent this?

Burgers and Beer Anyone?

The McDistance Map created by Stephen Von Worley depicts the mainland USA in terms of distance from the nearest McDonald's. As on a population density map, one can clearly see the east west boundary of population expansion. I think this map is also perfect in its simplicity -- no labels, streets or other markers are necessary. You can clearly see patterns of major highways, rivers and areas of rural or no population such as the Everglades and the Adirondack Mountains.

The Best Beer in America Map designed by Rick Lyke portrays the USA by state based on the number Great American Beer Festival gold, silver and bronze medal winners since 1987. Selected for its fun topic, this map has good and bad points. The colors used to fill in the states are in the same pigment and fade from dark (most wins) to light or white (least or no wins). The labels are simple and there is no data overload. However, this map does not consider the population size of each state and would look different with this additional data.