Graphic Narratives of Open Scholarship: a cartography of #FemEdTech


In alignment with Weller’s (2011, 2018) notion of the digital scholar, I collected and analyzed data sources from the open Internet. Social media plays an important role for scholars engaging in open scholarship as it provides a space to connect, collaborate, co-create, and disseminate knowledge (Quan-Haase & Sloan, 2016). As Stewart (2016) describes, Twitter is the ideal form of social media for open scholarship because users post without the expectation of privacy. These posts can be read and viewed by researchers without being logged into the platform. In many ways, the content posted by contributors is more authentic than other forms of qualitative inquiry because the researcher can observe phenomena without intervention (McCay-Peet & Quan-Haase, 2016). Pennington (2016) describes Twitter as a “naturalistic” environment for observation because participants express their opinions freely. Stewart (2016) also cautions that researchers must be careful about sharing public Twitter posts as there is a danger of context collapse and amplifying opinions not meant for a broader audience (p. 262). For this reason, display of tweets in this project are contextualized, where possible to ensure intent and meaning are clear.
There are hundreds of millions of users on Twitter and this medium creates numerous opportunities and challenges for researchers (Quan-Haase & Sloan, 2016). One challenge is the nature of diffused content on this medium. One remedy offered by Stewart (2016) is to research about network participatory scholarship. I wanted to focus on a much smaller subset of Twitter’s userbase, a loosely formed community within what Stewart calls “academic Twitter” (p. 252). This loosely formed group comprises scholars at all levels of their career development, from graduate students to academic teaching staff to part-time instructors to tenured faculty. Brown et al. (2016) describe that one way to collect data around an issue or community is through the use of metadata, keywords, and users’ profile information. For this research, a key term was used to collect data. Increasingly, this strategy is being adopted by researchers and is starting to get acceptance in the academic circles as there is a wealth of information without which data collection would be lost.

Data Collection

Within academic Twitter, there are a variety of hashtags that act as a form of organization of related topics. Using the hashtag, scholars have Twitter conversations and share open blog posts, articles, and various multi-media. I selected #FemEdTech as my primary hashtag for data source and subsequent analysis because of its intersections with feminist critique of educational technology and over-arching goal of practicing, thinking, writing, teaching, and learning in the open. 
All tweets under the hashtag #FemEdTech were collected between March 1 and July 5, 2019 and formed the basis of the study’s data analysis. The time period chosen was lengthy enough to get a full flavour of the kinds of discussions occurring to capture the essence of contributions made by the contributors to the group. This length of time was also approved by the committee through the proposal approval.   


Because the data sources are publicly available on the Internet and often released under an open license, materials are readily available for use and authors actively encourage reuse with attribution. Furthermore, the Brock University Research Ethics Board (REB) indicates that Internet data from open websites are exempt from ethical review board clearance, provided there is no staged intervention from the researcher (REB, 2012). Nevertheless, a formal application detailing the manner in which data was to be collected, stored, and analyzed was submitted and a formal exemption letter was issued by the REB as a form of clearance, as per REB file #19-268.

Data Organization

A dynamic spreadsheet collected all tweets containing the hashtag #FemEdTech between the specified time period using a social network analysis tool called TAGSExplorer created by Martin Hawksey (2011). The TAGSExplorer tool uses an application programming interface (API) to collect all tweets using the #FemEdTech hashtag into a Google spreadsheet. The API collects numerous data points about the tweet. The relevant data points for this project are the unique identifier, the username who posted the tweet, the content of the tweet, the time and date of the tweet posted, and a link to the tweet in Twitter (Hawksey, 2011). See Appendix B for the header information for all data collected.
On Twitter, popular tweets are endorsed through the retweet button. The spreadsheet displays retweets as duplicates. For example, the tweet shown in Figure 7 by @FrancesBell was retweeted seven times and appears in the spreadsheet in seven rows.
Tweets that receive more retweets assist in identifying strong themes as they imply resonance within the community. In accordance with the recommendation of Brown et al. (2016), the data were “cleaned” (p. 125) to remove unnecessary characters (e.g., retweets) that show up as RT and URLs. After cleaning the data and removing the duplicate tweets, there were a total of 841 original tweets.

Data Processing and Analysis

I manually identified and coded the tweets and then collapsed them into increasingly larger categories as per qualitative approaches (Creswell, 2008). Tweets that were sent in relation to the three values activities were often tagged with #FemEdTechValues in addition to #FemEdTech. These tweets allowed a closer analysis of the #FemEdTech network’s emergent values as they use these for the creation of their code of conduct (FemEdTech, n.d.-a). A total of 151 tweets were tagged #FemEdTechValues. Dozens of participants in the network contributed to these values activities which led to broader themes, which will be presented in chapter 4 and discussed in chapter 5. The process of collecting tweets, categorizing them, drawing the narratives, and refining the categories ended up being as iterative as the values activities themselves. At first pass of categorizing when treating all tweets as equal, there were many more thematic categories. Using the values lens to direct analysis allowed for a more focused interpretation of the themes. This iterative process will be discussed in greater detail in the self-reflection section of chapter 5’s discussion.

Scope and Limitations

Two major issues must be noted regarding the scope and limitations of this study. The date selection largely was arbitrary, based on convenience, while still preserving quality of the work. It was approved by the supervisory committee. There have been thousands of tweets since July 2019 and more accumulate daily. The events and online activity on Twitter subsequent to that date could not have been anticipated when the data collection was planned.
Secondly, it must be noted that anyone with a Twitter account can tweet using the hashtag #FemEdTech whether they are in alignment with the code of conduct or not. The volunteer rotating curators who manage the @FemEdTech account abided by the code of conduct on retweets, replies, and favourites. This is designed to limit the amplification of tweets that use the #FemEdTech hashtag that are not in alignment with the code of conduct. There is a real risk that a group or multiple groups could co-opt the hashtag for its own purposes to subvert messaging and counter the codes of conduct.

This page has paths: