Studying the ephemeral, cultures of digital oblivion Identifying patterns in Instagram Stories.

Introduction

While some argue that due to technology remembering was the default today, forgetting the exception (Mayer-Schönberger, 2011: 2) and, in a subtle revolution, remembering swaps with forgetting (Assmann, 2016: 205–208), Instagram stories turn this concept upside down, being available for 24 hours only. As 2021 marks the year of Germany’s federal election we’re curious to see how politicians and political parties use this ephemeral medium in a total-recall world of social media.

While Instagram use for past elections, e.g. the 2014 Swedish (Filimonov et al., 2016), 2016 U.S (Muñoz and Towner, 2017; Towner and Muñoz, 2020), 2017 German (Voigt and Seidenglanz, 2019) and 2018 Swedish election (Grusell and Nord, 2020), and similarly Justin Trudeau’s use of Instagram (Lalancette and Raynauld, 2019), have been investigated, stories have mostly been omitted.

We present results from our ongoing project, collecting and analyzing ephemeral Instagram stories. We collect stories by political parties, their front-runners and, additionally, stories by influencers and members of the public as baseline material of Instagram use. Bainotti et al. (2020) were able to outline two grammars for Instagram stories, one for interaction, one for documentation. Further they observed users to follow certain aesthetics and norms of influencer marketing. Their work serves as a guideline for our project to identify typical patterns of Instagram stories for each group, curious to see the aesthetics and norms used by politicians and parties. Our project is guided by the following research questions:

How do the prototypical Instagram stories of candidates and parties of the 2021 German federal election during the final phase look like?

How do their stories compare to other user groups?

How do ephemeral stories of politicians and parties compare to their permanent posts?

Data Collection

To answer ethical challenges (Bainotti et al., 2020; Franzke et al., 2020; McCrow-Young, 2021; Zimmer, 2010), we decided to acquire our corpus employing different methods for different target groups: a) political actors and influencers whose profiles are public, b) members of the public, whose profiles may not be public, and thus their stories are intended for a small group of people only, while they expect the ephemeral content to expire. To answer this challenge, we collected our corpus: a) using a custom-made scraper, b) crowdsourcing codings of stories through a browser plugin.

The first method is similar to Bainotti et al.’s (2020): We access Instagram using a custom script to emulate a user accessing the platform. For method b) we created a plugin for the chrome browser, allowing participants to code stories of the profiles they follow from within the browser. While method a) allows us to save and collect the corpus of stories as image or video files and their meta data, method b) only saves the users’ codings and some meta data. Scraping took place from Sep 13 through 26, 2021, counting 2208 stories. A first batch of crowd-annotations was collected in November, more codings using an improved plugin are planned.

Data Analysis

We are going to conduct a content analysis as outlined by Rose (2016: ch. 5), and used in previous work (e. g. Bainotti et al., 2020; Towner and Muñoz, 2020): The first step is going to be an initial open coding of the collected images by the authors, supported by codes from the literature. Inspired by Bainotti et al. coding will focus on the content as in “what does the image show”, as well as the composition, as in “which Instagram specific affordances were used” (e.g. polls, questions, countdown, gif, …). Further narrative style and context of use are added. Coding of stories by members of the public will be the most challenging, as our approach means we must rely on our crowd supporters’ annotations using predefined codes. They will be evaluated by coding predefined public profiles. Data analysis will start in early 2022.

Discussion

Our research expands the methodological framework for collecting ephemeral content on social media platforms. By introducing the annotation plugin as a bypass for ethical concerns collecting private Instagram stories, we hope to inspire future research analyzing personal content on social media, especially ephemeral formats. If successful, our analysis will improve our understanding of Instagram stories and their use by political protagonists during the hot phase of an election campaign. Future work could compare the content of stories and posts to the users’ content on other platforms like Twitter. Further we would like to train a model for automated coding and large-scale analysis.