Social media has moved from an occasional dissertation topic to a standard data source across multiple disciplines. Sociology, communication studies, political science, public health, marketing, psychology, and management all produce dissertations using social media data.

The methods are newer than most others in the research toolkit. The ethical questions aren't yet fully resolved. And the platforms keep changing the rules.

When Social Media Data Is Appropriate

Not every topic that mentions social media warrants social media data. Be clear about what your research question actually requires.

For those starting their research, argument structure calls for a different approach to the basics alone would suggest. The payoff comes when everything connects together, because each section builds on the previous one. Understanding this dynamic changes how you approach each chapter.

If you want to understand how people discuss a topic publicly (political discourse, health misinformation, consumer sentiment), content from public accounts may be appropriate primary data.

If you want to understand individual experiences, behaviour, or attitudes, interviews or surveys with social media users are usually better than scraping their posts.

If you want to examine organisational communication, corporate social media accounts are legitimate primary sources.

The Public/Private Boundary

This is the most contested ethical question in social media research. Legally public content (tweets from a public account, public Facebook posts, public Reddit threads) may not be ethically equivalent to content that's truly expected to be public in the way that a press release is.

A person who posts on a mental health support subreddit may not expect that post to be analysed in an academic dissertation. The fact that the content is technically accessible doesn't mean its use requires no ethical consideration.

Key Considerations and Best Practices

The ESRC Framework for Research Ethics (2015, updated 2022) addresses internet-mediated research. The key questions are: would participants reasonably expect their content to be used in this way? Does the research cause harm? Does the benefit of the research outweigh the privacy costs?

Your university's ethics committee will have a view on this. Apply for ethical approval before collecting any data, even from public sources, if your research touches on health, politics, or other sensitive topics.

Data Collection Methods

Quality over quantity. Always. A focused dissertation beats a sprawling one. Markers reward focus. They appreciate it. We help you stay focused. We trim the fat. We keep your argument lean and sharp. That's the goal. That's what we deliver.

Manual collection. Reading and copying relevant posts. Suitable for small samples. Time-consuming but allows close reading.

Twitter/X Academic API. Until 2023, Twitter offered a free Academic Research tier allowing researchers to collect large volumes of historical tweets. Post-2023 changes to the Twitter API have made academic access much more restricted and expensive. Check the current status before planning a dissertation that depends on it.

Reddit data. Reddit's API has also been more restricted since 2023. PushShift (a third-party archive of Reddit data) has reduced accessibility. Qualitative netnography or targeted manual data collection from specific subreddits may be more feasible for student dissertations than bulk automated collection. It matters.

Web scraping. Automated collection of publicly available content. Raises platform terms of service questions and requires technical skills. Most UK universities advise using APIs where available rather than scraping.

Surveys of social media users. Asking people directly about their social media use, perceptions, and experiences. More ethically straightforward than passive data collection. Suitable for attitudes and behaviours research.

Analysis Approaches

Quantitative content analysis. Systematic coding of content into pre-specified categories. Requires a clear codebook, intercoder reliability testing if more than one coder is involved, and transparent reporting of how categories were assigned.

Expert Guidance for Academic Success

Qualitative thematic analysis. Reading posts as text and identifying recurring themes. More flexible than quantitative content analysis; better suited to exploratory research questions.

Discourse analysis. Examining how social media language constructs meaning, identity, and social reality. Particularly appropriate for critical research on media framing, political communication, or health discourse.

Network analysis. Mapping relationships between accounts: who retweets whom, which accounts cluster together, how information spreads. Requires specific software (Gephi, NetworkX, Cytoscape) and technical skills.

Citing Social Media in Academic Work

Seeking feedback from peers as well as your supervisor gives you a broader range of perspectives on your work and often highlights issues that someone deeply familiar with your field might take for granted.

You need to cite the posts you're analysing, not just mention them.

APA 7th for a tweet: Author, A. [@username]. (Year, Month Day). Content of tweet in italics [Tweet]. Twitter. URL

APA 7th for a Facebook post: Author, A. (Year, Month Day). Content of post in italics [Facebook post]. Facebook. URL

APA 7th for a Reddit post: Username. (Year, Month Day). Title of post [Reddit post]. Subreddit. URL

Practical Steps You Should Follow

Always archive the URL using the Wayback Machine or Perma.cc at the point of collection. Social media content disappears. Posts get deleted. Accounts get suspended. If your primary data source disappears after you collected it, you need a record.

Frequently Asked Questions

Making effective use of tables, figures, and other visual elements can help communicate complex data more clearly than text alone, provided each visual element is properly labelled, referenced, and integrated into your discussion.

Q: Do I need ethics approval to collect Twitter data for my dissertation? A: Almost certainly yes if the content relates to sensitive topics, if you're identifying individual users, or if any of the accounts could belong to vulnerable individuals. Even for clearly public content on non-sensitive topics, your university's ethics committee should confirm the pathway. Apply for approval before collecting data.

Q: How do I anonymise social media data in my dissertation? A: The standard practise is to paraphrase quotations rather than reproduce them verbatim (since verbatim quotes are searchable and could re-identify the user), to remove usernames, and to change identifying details. Note that complete anonymisation of social media data may be difficult because the original post remains searchable. That's the honest advice.

Q: What's netnography and is it a valid dissertation methodology? A: Netnography (Kozinets, 2010) is an ethnographic methodology adapted for online communities. The researcher immerses themselves in an online community over time, observing interactions and participation norms, with full disclosure to community members. It's a legitimate qualitative methodology, particularly used in consumer research and social science. Your ethics application will need to address how you handled disclosure and consent within the community.