Through the library at CSU Pueblo, Dr. Rees was able to have Historical GIS: Technologies, Methodologies, and Scholarship bought for us by the fabulously clever librarian Karen Pardue. Written by Paul Ell and Ian Gregory, Historical GIS offers up a comprehensive overview of GIS and its applications in the field of history. It’s a fairly technical book but its insights into how GIS can be used in history and also its strengths in limitations has already proven to be enormously helpful in how I find myself thinking about our GIS project. I will attempt to share what I’ve learned so far so when I talk about GIS, there is some more context to it then just “oh it’s the software that’s going to make all of this data shiny and pretty and map-like.”
The first question is, and should be, what is GIS? Well, broadly speaking, GIS is a type of software that allows the input of spatial data and attribute data into a database that can be used to project visual presentations of the data. Most importantly, that data can represented spatially over time which is where the inherent usefulness of GIS to a historian becomes apparent. attribute data describes what a feature is, and spatial data is the attribute data is inputted and linked to information about where the feature is located. This spatial data is stored as co-ordinates. More narrowly, GIS is the tools that the program offers, in addition to the software that provides the tools.
For our CF&I project, we will be essentially using three components of data: spatial, attribute, and time (known as temporal data). These three data groupings will allow us to visual the impact of change over time and space to create an understanding of movements of Hispanic workers. In an ideal situation, all three components of data would be readily available. However, since historians almost zealously maintain that their field is qualitative as opposed to quantitative, there are gaps because historical data is sometimes limited.
So here’s where thinking about data as projections spatially that have a quantitative approach gets tricky. We have fixed, limited data for Hispanic workers of where they were, and when they were at that place. So these three points of data need to be categorized much like you would in a scientific experiment; one group of data needs to be a fixed component, one group needs to be a controlled component, and one group of data needs to be the measured component.
This is just an example of how you can organize the data. For GIS, Chris and I go back and forth on what we think could be the best way to organize the data. Our arguments are very philosophical in nature because the truth is, we could very well be putting the cart before the horse.
So here is what Chris thinks the data should be:
|GIS Project Input||Hispanic Miners||1920s||Southwest|
His argument is that we are measuring where the Mexican workers are going in the 1920s.
My table looks like:
|GIS Input||1920s||Southwest||Hispanic Miners|
I believe that we are examining the movements of Hispanic workers within a fixed time period in a spatial area, and we are measuring where the Hispanic workers are moving within that area.
What these tables really show is what Dr. Rees considers to be the best part of digital history. You begin to ask questions that would never come up in a traditional historical educational settings. You begin to think about source compilation, and the best way to share the data you have compiled. This is one of the reasons this blog post has taken so long to complete. In a way, I feel like this whole ordeal has rewired my brain.
So at the end of creating a GIS project, the goal is to have an organization of data that gives the ability to visual said data. The results gained from analysing this spatial representation is called spatial analysis. Ideally with a project with project like our GIS project, one could integrate two databases together to find causational relationships. The thing to keep in mind though, is that the map is not the end goal for a GIS historical approach, it is a research tool.