Oath Keepers by Sam Jackson

Oath Keepers by Sam Jackson

Author:Sam Jackson
Language: eng
Format: epub
Publisher: Columbia University Press


Data Processing and Analysis

After creating local archives of the website and blogs, I used software to extract the text from each page.12 This tool allowed me to set aside information like page headers, banners, and other items that appear on each page. For example, each page on the Oath Keepers website contains a sidebar with links to external sites and brief descriptions of important Oath Keepers documents and individuals (such as their “Declaration of Orders We Will Not Obey” and their board of directors). Using this tool, I ignored this text and only extracted the body of the content on each page. In addition, I excluded comments left on each page, most of which are written by visitors to the site. Social movement scholars have argued that both elites and nonelites participate in discourse that helps social movements make sense of their world; importantly, the results of framing analysis may differ dramatically depending on which group is the focus of analysis.13 For this study, I focus on content produced or shared by Oath Keepers rather than content produced or shared by individual supporters (or opponents) of the group. After excluding repeated content and comments from readers, I am able to get a more accurate sense of the amount of textual data that I have. This process also allows me to use automated text analysis to get an overview of my data.

Video data poses a greater challenge for analysis. My analysis in this project focuses on text. For videos, that means analyzing what those who appear in the video say, setting aside characteristics of the audio (for example, voice pitch or speaking speed).14 I mostly set aside images that appear in videos as well, though I include some particular images that serve an important function for Oath Keepers.

To facilitate analysis, I created transcriptions of each of the 136 videos collected for this project. Most videos (99, totaling nearly 29.5 hours) I transcribed by hand. Thirty-seven videos I transcribed with Trint (http://www.trint.com), an automated audio-to-text service, totaling just over eight hours of content. For each video automatically transcribed by Trint, I watched the video and corrected the transcript where necessary.

In total, the data used for this project (website pages, blog posts, and video transcripts) contain approximately 1.4 million words, or more than 2,500 single-spaced pages. Given this large amount of data, I used automatic text analysis to create an overview of the data. I created a list of more than 400 keywords that signify the presence of certain topics and actors; several keywords also identify types of documents (for example, testimonials and documents originally published on a different website).15 I use this list of keywords to identify eight topics (American history, people, political issues, P/M movement, gear and tactics, political system, conspiracies, and miscellaneous) in the documents; many of these topics also contain subtopics (see figure A.1). Each document may contain more than one topic and subtopic.

Table A.1 shows a breakdown of how many documents mention each of the main seven topics (excluding the miscellaneous category).



Download



Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.