Every dissertation lives or dies a little on its sources. Cite the wrong kind — outdated, low-quality, or simply irrelevant to your specific argument — and even a well-written chapter starts to feel shaky under a supervisor's eye. Here's how to actually find sources worth building an argument on.

Start With Databases, Not Google

General search engines are built to surface what's popular, not what's rigorous. Academic databases — Google Scholar, JSTOR, your university's own library portal, and field-specific databases like PubMed or IEEE Xplore — are built to surface what's peer-reviewed. That single filter removes most of the noise before you've read a single abstract.

Check the Publication, Not Just the Title

A paper's title can be perfect for your argument and still be a poor source if it was published in a predatory or low-standard journal. Before citing anything, check: is the journal indexed in a recognised database? Has the piece been cited by other credible work? Is the author affiliated with a real institution?

Read Backward From Recent Papers

One of the fastest ways to build a source list is to find one strong, recent paper in your area and work backward through its own reference list. Recent papers have already done the work of identifying which older sources still hold up — you're borrowing their judgment, not just their bibliography.

Keep a Running Annotated List

Don't just collect PDFs — for every source, write two sentences the moment you read it: what it argues, and how it might fit your chapter. Six weeks later, "I'll remember why I saved this" is rarely true.