This project was created as a final project for an introductory course on digital history for Professor Graham at Carleton University. The guidelines for this project were:
To work with a set of data whether it be a set provided or a set put together yourself.
To analyze this data using relevant and suitable tools and techniques introduced in the course.
To visualize the results of the analysis of the collected data using the tools and techniques introduced in the course as inspiration.
My implementation plan was to follow these guidelines to the best of my ability according to the topic I chose within an area of my research interests. Overall, the plan was to keep things simple and incorporate as much analysis and visualization as was appropriate for my data, and within the time constraints of the project due date.
The main goal of this project was one that was both exploratory and experimental in nature. The general aim was to explore the ways in which the tools and techniques for conducting digital history could be applied to a research topic more traditional in nature. By "more traditional in nature" I am referring to the idea that the topic for this project would typically lie within the the academic streams of urban and social history. In other words, I have applied digital history methods of collecting a large dataset from an online archive, using the tools of the digital humanities to analyze that data, and using the visulaiztion tools covered in the course to present the results produced from the analysis of the collected data.
In this context the methods used are a combination of typical historical research conducted through the consideration of primary and secondary literature concerning the subject matter, and the use of tools and techniques generally applied by those in the digital humanities for dealing with very large datasets. In my opinion, the end result was a sort of hybrid of these two methods of conducting historical research, each serving to compliment the other. In other words, traditional research methods were used to address very specific, narrow questions such as those surrounding the compulsory vaccination debate and questions of urbanization and public sanitation. Digital history techniques were used to compliment this by allowing for a broad, overall look at the severity of infectious disease epidemics by city and date over a period of 100 years.
Research sources used for this project are very typical in nature. Primary sources, for example, consist mainly of historical newspapers available from the BANQ collection. 19th century medical journals and other sources of this nature were collected automatically by using programs and methods covered in the course from the Canadiana Discovery Portal.
Secondary sources used are also rather typical in nature, and consist mainly of academic journal articles and peer reviewed papers; some being available online as public domain and others available through MacOdrum Library at Carleton University.
Other sources used for very general information include Wikipedia and the World Health Organization website.
All documents used in the research concerning this project are linked to within the project itself, with the exception of those which are not public domain, where a basic reference method is used to indicate the name of the author, the title, and date of publication for others who may be interested in further reading.
As stated above, the use of computer-based visualizations in this project are meant to add a level of depth to the more traditional research done by myself. All methods are meant to be rather basic for the sake of readability and to clearly convey the results of the analalysis of the collected data.
All other dosumentation used to build this project such as datasets, graphs, and html pages can be found in the gh_pages branch of my GitHub repository.
As mentioned, this project and all accompanying material is hosted on GitHub.
All accompanying pages are written in html and tested using Google Chrome; all visualizations are in .png format and created using Microsoft Excel and open source visualization and topic modeling tools introduced in the course.
This project, all accompanying data, files, and tools used are open access and public domain with the exception of Microsoft Excel. All primary sources of research and the online archive from which the datasets were created are public domain. Academic secondary research sources are the exception to this in terms of access; most of which are in the form of academic journal articles.
As explained in the project itself, the major limitation of this project lies within the fact that only one online archive was used to collect relevant data. As such, the visualized results of the collected data reflect only the sources contained within that archive, and in this context are limited in scope.
Another limitation of this project is that it considers the presence and frequency of publications concerning the diseases covered as indication of the severity of their spread and presence, where a lack of published material in a particular city at a particular date does not necessarily indicate an absence of the infectious disease in question.
Another question that was initially intended to be, but utlimately not addressed in this project was a consideration for mortality rates due to infectious disease by date and city and according to relative urban popualation. Ultimatley, time constraints and my relative inexperience in conducting this type of research restricted my ability to approach this topic in this level of detail within the time limits of the project's deadline for completion.
One other limitation of this project was in the fact that many of the texts collected from the archive did not contain publication cities and dates. The consequence of which was that the datasets I assembled addressing this question were considerably smaller in size. While this was originally a major concern of mine, I percieved the size of these datasets to be sufficient enough to carry out an analysis of this type and was pleased with the results.
At the start of this project I was critical of how accurate the results of the data analysis would be according to primary and secondary literature on the spread and severity of the infectious diseases considered. I had previously studied the smallpox epidemic in Montreal between 1875 and 1885 in the context of Canadian scoial history and the French versus English culture clash created by the debate surrounding compulsory vaccination. As such I had some previous knowledge of the severity and spread of this disease and was not convinced that using digital history methods to analyze the data collected would produce results that reflected what my previous research had indicated. In this sense I made sure to frame this project as one that was experimental in nature, to see whether or not, or to what extent, an analysis of the data collected would reflect what my research has indicated in terms of key dates and topics within contemporary published literature.
I was quite surprised to see that the results of the data analysis coincided with many of the conclusions drawn in my previous reasearch, particularly considering the cities and date ranges of major epidemic outbreaks. What also surprised me was the relative accuracy of the very basic level of topic modeling that I had conducted on this data, where major issues like the transmission of cholera through contaminated water, and the concerns surrounding smallpox vaccination were clearly indicated.
In this sense, where I had fully expected to argue that this type of research was not appropriate for dealing with material of this subject matter, I was pleasantly surprised to be able to argue the opposite. The end result for me is in the fact that this type of approach to conducting historical research is one that I will be able to consider in future projects as a means of adding depth and scope to my research.