Tuesday, March 6, 2012

Who Were These People? GIS and Wiawaka's Visitors

Back in January, I introduced GIS as a tool to help understand Wiawaka's history.  I will use this post to give some concrete examples of how this works.

Last summer, when I had the opportunity to do preliminary documentary research, I recorded some of the guest books. I wanted to explore who the visitors to Wiawaka were, and the guest books include information like the person's name, address, religion, how much they paid for their room, and how long they stayed. This information, coupled with information available in the census for their home-town (who they were living with, age, color, occupation, whether they owned or rented their home, whether they were literate or not, etc.) can paint a fairly detailed image of who visited Wiawaka. I used the sample data I collected to see what sorts of information GIS would reveal. Click on the images to make them larger and make the data easier to read.

The data I had to work from was a sample of one hundred guests who visited Wiawaka in 1931 (see Figure 1). Unfortunately, although the 1930 census has recently been made available, it has not been completely indexed for searching. New York is one of the states that has not yet been indexed, and I did not have time to manually search the census for individuals living in larger cities like New York City! So, I could not incorporate the census data in my trial run.

Figure 1: Sample entry page from a Wiawaka guest book. Courtesy Wiawaka Holiday
House Archives, Rensselaer County Historical Society, Troy, NY

Loosely using Fyfe and Holdsworth's (2009) study of small-town hotel guest registers from other locations in New York state, I ran analyses of the home residence data of 100 of Wiawaka's 1931 guests. The very limited sample that I used provided some interesting suggestions about Wiawaka's guests in the early years of the Depression. Please note: these results are extremely preliminary! More in-depth analysis is needed to draw any conclusions, but I did see enough with this data to convince me that I will be using GIS in my analysis. Here are a couple of examples of the analyses I looked at.

Using GIS, I mapped the location which each of the 100 sampled 1931 visitors listed as their home residence in the guest book. While each location shows up only once on the resulting map, each location may (and often does) represent more than one visitor.

Figure 2: Where visitors came from, and their 1931 "Center of Gravity"

1) Measuring the visitors' "center of gravity." I asked the GIS program to calculate a value known as the central feature of the points, or locations, that I mapped (see Figure 2). This shows the average location of all the points, taking into account both distance from each other and the number of individuals from each place (a location with many Wiawaka visitors, therefore, would have more influence or "pull" on the central feature than a location with only one Wiawaka visitor). The central feature for the sample of one hundred 1931 visitors was located just south of Troy. This, in and of itself, is not terribly interesting. However, by calculating the central feature for many years worth of Wiawaka visitors will show if, and how, the geographical demographic changed. For example, if the central feature for the 1920 visitor data is located further south than that for 1930, it will indicate that, in 1920, more people visited Wiawaka from New York City and vicinity than from Troy. To me, this data is interesting if it shows shifts and if it shows no shifts -- the implications of a changing demographic or one that stays the same are equally interesting.

Figure 3: Proximity of visitor residence to Wiawaka.

2) Proximity of visitor residence to Wiawaka. I used the GIS program to draw rings around Wiawaka at 50-mile increments so that I could see how far people were traveling to visit (see Figure 3). While Fyfe and Holdsworth found that most of the visitors to their hotels came from within a 50 mile radius (2009), this was true only for 30% of the sampled Wiawaka visitors. Curiously, there was a concentration of visitors coming from the ring representing a distance of 150 to 200 miles from Wiawaka, including people from New York City and northeastern New Jersey; Scranton, Pennsylvania; and Ithaca and Rochester, New York.

Figure 4: Mapping Wiawaka visitor residences and railroad lines.

3) Influence of the Railroad? Changes in transportation routes and technology have been used to explain changes in visitor patterns at sites, including hotels. These include road improvements, railroad lines, canal access, and later in the twentieth century, changes in automobile technology. To see if transportation routes may have played a role in the pattern of visitors to Wiawaka (especially perhaps to explain that large group traveling between 150 and 200 miles) I plotted the home residences of Wiawaka visitors on a map showing rail lines (see Figure 4). Interesting: all of the 100 visitors sampled from the 1931 guest register lived in a town or city along a railway line. The 150 to 200 mile distance from Wiawaka may represent the distance/travel time that visitors are willing to travel by rail to spend a few days at Lake George.

Fyfe, David A. and Deryck W. Holdsworth (2009) Signatures of Commerce in Small-Town Hotel Guest Registers. Social Science History 33(1): 17-45

No comments: