DeMaggio- Week 2

Chapter 1

In this first chapter, Mitchell introduces GIS analysis as a whole, explaining the process one needs to follow to use GIS programs like ArcGIS efficiently and the best way to present your data. You need to know what question you’re asking, what information is required to answer your question, understand the data you have as well as choose a method to map your data that represents your findings the clearest. Your results can either be mapped as discrete or mapped as continuous, or even mapped as a summary of areas. However, while all three approaches are using the same data, the end results of both methods will be different, making it important for you to choose the method that will convey what you’re presenting the clearest. Mitchell also walks us through the differences between vector models and raster models, saying that with vector models, “each feature is a row in a table, and feature shapes are defined by x,y locations in space”, and those areas are defined by borders and are represented by closed polygons. From there he explained that with the raster model, locations aren’t defined by specific coordinates but rather with matrices of cells in continuous space and that the sizes of the cells can be altered to fit the data that you have, which he goes further in-depth later on. From there he lists different types of attribute values such as categories, ranks, counts, amounts, and ratios. Categories are groups of similar things, ranks put features in order from high to low, counts and amounts show total numbers, and ratios show the relationship between two quantities. This was a lot of information to try to retain, especially for this being my first time diving deep into GIS analysis, but in the following chapters, each definition and feature are further explained, aiding me in my understanding of it all.

Chapter 2

Chapter 2 talked about mapping where certain things are, such as crimes, businesses, employees, etc. When asked the question of “why” it seems obvious for the reason to look at a map is to see where a certain feature is. While that still remains true, mapping where things are helps make patterns noticeable and from there you can decide where you need to take action. An example of this in the textbook is when you map where certain crimes (burglaries, theft, auto theft, assault) occur in a specific area. From there you can see where they all have occurred, and see patterns in where there are more crimes in one area in another, as well as where certain crimes are more likely to occur. This is made possible by starting with a basic map with all of the same symbols, from which you can move to divide the feature into different categories, making your data points more specific by either using different symbols or different colors. Mitchell then dives into how you use your map and states that it is paramount when creating a map, to make sure that the map is appropriate for the audience you’re addressing as well as the issue that you are addressing. If your audience isn’t familiar with the area you’re representing, it’s good to add reference locations such as major roads, lakes, or administrative boundaries to provide more context to your map. Not only are reference locations important, but so is the amount of categories you decide to use. If you use too many categories the patterns in the map can become too complex to see, however, if you include too few categories, essential information can be lost. The same thing goes for symbols, it is easier for people to discern between different colors than different symbols if there are enough points of data that are clustered together. The end goal for your map is to convey the patterns and information you desire in the clearest and most efficient way possible. 

Chapter 3

This chapter focused primarily on quantitative data and was by far the most when it comes to information overload for me. Early in this Mitchell states that mapping the most and the least allows you to compare places based on quantities, which can help bring out patterns and a better understanding of the relationships found in your data. From this basic understanding, the next step is to understand the three features and what they each entail, where discrete features are individual locations, linear features, or areas, continuous phenomena as defined areas, and data summarized by shaded areas. The theme of the audience is revisited in this chapter and the discussion of how the appearances of maps differ between the exploration and the presenting of the data you study. If you are simply exploring your data, then your map should be more detailed as well as mapped in various different ways. If you are presenting the map and data, your map should obviously be more specific with the relationship you’re attempting to prove to your audience. From there the chapter dives back into the different attribute values, where I learned more about the use of ratios. Ratios in this chapter were very important when it came to displaying the highest and lowest values of data, and especially important when it comes to shaded areas on a map. Ratios help generate the differences between large and small areas. This can be especially useful when it comes to finding proportions and densities, which are talked about in chapter four. Counts, amounts, and ratios are usually grouped into classes because each feature in your map can have different values, especially when the range of values you have are larger. When creating classes it’s important to know where each feature will lie in your classes, because if you change the classes of your map, the map can look very different from the one before. We then go into the different kinds of class breaks. Natural breaks are where classes are based on natural groupings of data values, quantile is where each class contains the same amount of features, an equal interval is where the difference between the high and low values is the same for every class, and standard deviation is where features are placed in classes based on how much their values vary from the mean. They all operate differently, meaning that they all have their advantages and disadvantages which can make it difficult to decide which class type will be most effective in appropriately displaying your data.

Chapter 4

Chapter four is completely on map density, which in the beginning states what it is and what it does: mapping density can show the highest concentration of a feature you’re examining, it can be more efficient than just mapping locations, and it’s good for census tracts and counties. However, there is a difference between the two methods of mapping density, by defined area and by density surface. When you go by a defined area, you can use a dot map or calculate a density value for each area, which allows you to see density graphically. When calculating the density value of each area, “you divide the total number of features, or total value of the features, by the area of the polygon” and from there each area is then shaded based on its density value. When mapping density value it’s best to use different shades of a color, typically the lightest shade indicating the lowest density and the darkest shade representing the highest density value. When mapping by density surface it is usually created in the GIS as a raster layer, which we learned in chapter one as matrices of cells in continuous space. The benefit of mapping by density surface is that it provides the most detailed information in comparison to mapping by defined area, however, requires more effort to do. It was nice that there were tables included in this chapter that stated when to use one or the other; you should map density by area if you have data already summarized by area, or lines/points that you can summarize, and you should map density by surface if you have individual locations, sample points, or lines. I feel that most of the time when mapping density it would make more sense to map the densities by shaded areas rather than graphing dots because, for me personally, it’s easier to distinguish the difference between color shades as opposed to clusters of dots, because the clusters can then further skew the true value of density being portrayed on your map. Cell size and search radius also play roles in how your map and presented patterns appear. If your cell sizes are smaller, you’ll have a smoother display, and if your cell size is larger you’ll get a coarser image. The typical range of cell sizes to use is between 10 and 100 cells per density unit. With a search radius, the larger the search radius, the more generalized the patterns in the density surface will be, while a smaller search radius shows more local variation, but you have to be careful because if your search radius is small enough, most cells will have very low-density values, creating, “broader patterns in the data may not show up.” In all of this reading, I have learned that many factors in GIS mapping are a range or a scale, and it’s up to you to find the right proportions that will bring the most fruition to your map.