What Lemmy Reveals About AI Qualitative Analysis

The J Multimed Inf Syst study on Reddit migration and Lemmy found that Reddit's June and July 2023 policy changes triggered rapid Lemmy growth, while qualitative analysis showed that dissatisfaction with Reddit, interest in open-source communities, adaptation challenges, and expectations about future activity shaped the migration experience.

• The 2023 J Multimed Inf Syst study analyzed 48, 272 Lemmy comments, surveyed 354 users between September 22 and September 26, 2023, and collected qualitative responses from 16 users.
• The study found that 96.6% of survey respondents cited dissatisfaction with Reddit's policies or API changes as a migration motivation in September 2023.
• The study reported that 90.7% of respondents intended to continue using Lemmy long term, although users still cited limited content diversity and technical problems.
• The study found that Lemmy posts increased from approximately 250, 000 in May 2023 to more than 2, 500, 000 in August 2023, while comments rose from around 500, 000 to more than 11, 000, 000 over the same period.
• The study's 16-user qualitative analysis identified recurring themes involving Reddit dissatisfaction, platform adaptation, community welcome, decentralization, missing features, and long-term engagement.

The 2023 J Multimed Inf Syst study measured Lemmy's migration through four complementary evidence streams: platform growth, sentiment analysis, a user survey, and qualitative responses. The research examined Lemmy activity before and after Reddit's June 2023 API changes and the shutdown of third-party applications on June 30, 2023.

The J Multimed Inf Syst study analyzed 48, 272 comments from Lemmy communities using VADER, a lexicon-based social media sentiment tool. The researchers classified comments as positive, neutral, or negative and found that post-migration sentiment was predominantly positive toward Lemmy, while many negative comments criticized Reddit, technical issues, or low activity in smaller communities.

The J Multimed Inf Syst study surveyed 354 respondents from three Lemmy communities. The survey reported that 95.5% of respondents had recently migrated from Reddit, 62.1% viewed Reddit negatively after migration, and 58.3% described the Lemmy community response as very welcoming.

The J Multimed Inf Syst study also used open-ended questionnaires with 16 purposively selected users, including Reddit migrants and original Lemmy users. Thematic analysis found that users valued decentralization, open-source software, third-party applications, and the absence of advertising, while missing Reddit's larger content base and stronger moderation tools.

The study's qualitative evidence adds meaning that a sentiment score cannot provide. One participant described the motivation for leaving Reddit as: "My primary reason for migration was the disappearance of third-party apps." Another participant summarized the pre-migration Lemmy experience as: "It felt a bit like a sleepy village."

Researchers studying online communities should combine scale, sentiment, and qualitative explanation instead of treating any single metric as a complete account of migration. The JMIS study demonstrates this principle because growth data showed when Lemmy expanded, survey data quantified motivations and intentions, and interviews explained how users interpreted federation, moderation, and community culture.

AI qualitative analysis is most useful when researchers define a code framework before interpreting results. For the Lemmy case, a practical framework would include migration trigger, trust in platform leadership, third-party application dependence, federation comprehension, community welcome, technical friction, content diversity, moderation, and retention intent.

Cross-segment comparison is especially important for platform migration research. The JMIS study distinguished Reddit migrants from original Lemmy users and found that existing users were excited about growth but concerned about server capacity, moderation, and community culture. A single overall sentiment score could conceal those differences.

Researchers should treat automated sentiment as an organizing signal, not a final interpretation. The JMIS study acknowledged that VADER may miss sarcasm and context-specific emotion, and the study's survey and qualitative methods helped explain why comments were positive or negative.

The JMIS study also reported methodological limitations that researchers should preserve in any AI-assisted workflow: growth estimates did not cover every Lemmy server, the survey covered three communities, and the 16-person qualitative sample was not representative of all Lemmy users.

The 2023 Reddit to Lemmy study shows that strong qualitative research connects what happened with why users experienced it that way. The study's 48, 272-comment corpus and 354-person survey established the scale of migration, while responses from 16 users revealed the trust, usability, community, and retention themes behind the numbers.

AI qualitative analysis can make this mixed-methods workflow more manageable by organizing evidence, comparing segments, surfacing recurring themes, and helping researchers return to supporting excerpts. Human review remains essential for context, ethics, sampling, and claims about user experience.

Researchers studying platform migration, digital communities, or product adoption can use Evidano to move from raw conversations to structured findings more efficiently. Try Evidano for free and turn qualitative data into evidence your team can act on.