Dataset

Since I seek to discover broad trends throughout the United States, newspapers are one of the best mediums through which to analyze the spread of information on both a local and national scale. The main source is Chronicling America, which is a website that hosts a large database of American newspapers from 1789 to 1963. The search term I have used for this research is “Thanksgiving Indian.” It is extremely important to note that this is an offensive term for American indigenous communities. However, these newspapers are from a time where the term “Indian” was frequently used as a replacement for “Native American” despite the fact that it had been known for many years that Columbus mistakenly believed he had landed in India instead of the Americas. Having to use the word “Indian” instead of a more appropriate and respectful phrase already lends valuable insight into the types of depictions of Native Americans that will most likely populate the search results. 

Note: The derogatory word r-skins frequently appears in the corpus. This along with “indians” were often used interchangeably to reference Native Americans. However, I will not be focusing on this specific term, which warrants future examination as to whether which term appears more and if there is a regional difference in usage.

From these results, there are two sets of data I will be considering. The first is the data of the newspapers themselves including:

  • Location of the newspaper publication
  • Date of publication 
  • The name of the newspaper
  • How frequently a newspaper appears in the dataset

The second set of information is the most critical to my analysis and consists of the texts of the newspaper.