How to Actually "Do Your Own Research"
An editor's guide to fact-checking and fact-finding.
Contributor Jacob Bielecki has a new piece out in Queer Majority: “Pope Francis: The Surprising Yet Flawed Ally.”
Take the collapsing public trust in institutions, the press, and experts, combine it with ubiquitous smartphones and high-speed internet, and add a dash of DIY thinking, and the result is the “do you own research” era in which we now live.
This new ethos has been much maligned by the kinds of people who inspired the backlash it grew out of, but it’s a bit bizarre for any educated person to come out against research. In the current political climate, telling people, “Don’t do research, just trust us” is so rhetorically counterproductive that I have to question the intelligence of anyone who utters it. Of course, “do your own research” is more than a little conceptually muddled itself. The idea that any rugged individual can single-handedly recreate the function of knowledge-producing institutions is disproven by the fact that all of the information one finds when “doing their own research” is either generated by these very institutions or turns out to be unsubstantiated hearsay. In the end, every DIY researcher is really just a DIY curator, picking and choosing which institutional knowledge to trust, usually based on whether it conforms to their biases.1
Putting that aside, there’s a more glaring practical concern: most people don’t know how to research. Sure, nearly everyone has access to physical libraries, and we all know how to type a query into Google and hit enter. But those are only precursors of research. Beyond that, many seem unsure of or overwhelmed by the process. Public discourse is awash in postgrads with smart-person glasses insisting that the average citizen lacks the means to do proper research. This simply isn’t true. Having edited and fact-checked scores of articles written by academics or people with letters after their names, I know firsthand that formal training and institutional resources are no guarantee of good research. If you’re reading this, you have everything you need to research almost any issue. Advanced degrees and expensive tools are great, but lacking them by no means dooms you to darkness. These are the tips and tricks I’ve learned as an editor, writer, and fact-checker that anyone can make use of.
Presearch
Let’s start with some basics. Search engines are often the very first step one takes in beginning their research. Google specifically should be your search engine of choice. The others, such as Bing and DuckDuckGo, so seldom turn up better or more useful results that I long ago stopped bothering with them. I find myself turning to them only in rare moments of desperation — and they never help. Stick with Google, and make use of its filtering and advanced searching tools to hone your searches.
There are also some useful online practices I like to think of as “presearch.” If you’re at all like me, you’ll regularly come across something in the wild — an article, social media post, video, podcast, poll, or study — and then months later find yourself researching a related topic and trying to find it. Only, now you can’t. Like digital car keys, they aren’t where you recall leaving them, or where they’re supposed to be. You don’t quite remember where you saw it, or how the headline was worded, or what exactly the reported stat was. The web page may have also been moved, renamed, altered, or taken down. Now, search engines, social media, and the Internet itself are conspiring to gaslight you into almost believing that maybe it never existed in the first place. Maybe you’re just losing your damn mind. This is where presearch comes in handy.
Find a couple of good, web-based bookmarking sites that have both mobile apps and browser extensions and set up an easy system that allows you to one-click save links that may be useful in future research as you encounter them. I use Instapaper for written articles and Raindrop for studies and social media posts. Even as they pile up, searching through your bookmarks is far easier than searching for a needle in the haystack of the Internet. Most social media platforms have some form of bookmarking available, and that can be useful too, but if a post is removed, or the account that posted it is deleted, it disappears as though you’d never saved it. Saving the URL with a bookmarking client often allows you to retrieve a post from the dead even if it disappears from the site (more on this later). Get in the habit of saving stuff before you need it and can’t find it.
Getting Pointed in the Right Direction
Depending on the query, search engines may direct you to one encyclopedia or another within the first few results, and that’s never a bad place to start, especially with a subject matter you’re not deeply steeped in — and sometimes even if you are. The big three are Britannica (edited and reviewed by experts), Wikipedia (open source), and Encyclopedia.com (online aggregator), and they can all be useful resources. Yes, even Wikipedia, notwithstanding the scathing criticism we’ve published at American Dreaming. While Britannica is the only of the three that is credibly citable as a source on its own (though it’s never a good look to lean heavily on it), all can provide a decent familiarity with the broad strokes of most topics. Wikipedia in particular, due to its format, individually cites nearly every claim made in footnotes. Mining wiki footnotes can be a good way to find sources that search engines won’t find. Online encyclopedias are not proper research in and of themselves, but they can help give the early stages of your research some orientation.
Artificial intelligence, though far more versatile, should be seen in a similar light to Wikipedia. An increasing number of people use AIs to do their research, and I must strongly caution against this. The Internet is, and always has been, mostly bullshit. As long as AI is trained on the Internet, it will never be a reliable narrator, information source, writer, etc. Not to mention its propensity for hallucination and making things up. It can still be enormously helpful, but it’s not a magic information genie. Think of AI as a knowledgeable but non-expert friend you ask for advice. You can’t take your friend’s answers as gospel, or cite their words as a source, but their recommendations can nevertheless be useful for pointing you in the right direction.
Most AIs also have some kind of “deep research” function where they can perform a more rigorous dive into a subject and synthesize a much longer and more in-depth report than a standard reply. Again, very neat stuff, but file this in the “giving me ideas for what to look more into myself” category, not “this is everything I need to know.” If you’re wondering which AIs I use: mostly ChatGPT, regularly Perplexity, and occasionally Gemini, Grok, and Claude. Perplexity in particular is very good at AI web searches.
Asking Experts Directly
Research feels like a solitary endeavor, but reaching out to an expert who knows far more about a subject than you can be one of the most helpful steps you take. This is a common practice among journalists, one I’ve often used myself. My professional contact list has accumulated quite a number of experts over the years, and I’m not shy about periodically touching base with a question, seeking guidance, or asking for a quote. Bully for me, right? But this isn’t an avenue open only to journalists. Most people know some experts, either through their family/friend network, their work, or other associations. Moreover, there’s no harm in politely approaching people online, through private messages or preferably email, and asking for a moment of their time. I occasionally receive such messages myself (usually about LGBT issues, which I’ve covered for years through my work outside of Substack), and I always try to respond.
The moment it takes to reach out to someone, and the slight awkwardness that may entail, could save you many hours or even days of time. Whatever the subject of your research, there are often hundreds or thousands of people online who have dedicated their career to it. It’s beneficial to you to get their input, and flattering to them to be asked for it. A true win-win!
Articles, Essays, Studies, and Papers
Whether through Google, AI, social media, footnote-mining, or any other mode of discovery, most of the sources you’ll be connected with fall into two categories: journalistic (news reporting, articles, essays, and op-eds) and academic (scholarly papers, reports, working papers, white papers, preprints, peer-reviewed articles, and studies). When it comes to journalistic sources, there are two important considerations. First, depending on the topic at hand, you’ll usually want the most up-to-date information. The press covers large news items with many articles over a period of time. Often, the earliest articles have the least complete information and the most inaccuracies. Always checking the date of the articles (a more common mistake among writers than you’d think) and reading a handful of coverage pieces are useful filters here.
Equally important is vetting the credibility of the media outlet. Unfortunately, there’s no easy codex I can give you on which sources to trust and which not to. Once trusted sources sometimes lose credibility, smaller or less reputable outlets can establish themselves as more respectable, and publications rise and fall every day. And even questionable or highly biased outlets can be trusted in specific circumstances — such as when they report on something that cuts against their own slant or when they echo the reporting of outlets with very different biases. If the Washington Post and Breitbart agree on something, that’s a very good sign.
Some media organizations are also credible on certain subjects but not others. The New York Times, for all that it’s denigrated by conservatives, is the long-standing newspaper of record in the US and one of the most credible media publications in the world. But from perhaps 2019 to 2023, much of their coverage of various cultural issues, such as trans politics, was objectively low quality, riddled with inaccuracies and distortions, and seemingly catered to activists rather than the truth. The veracity of their coverage on the issue has improved dramatically in the years since. Picking up on these wrinkles is a product of experience. It can’t really be taught, because the landscape shifts too much. The more you research, the more finely attuned you’ll become. (For more on media literacy, see this article I wrote on the problem of politicized fact-checking.)
For academic sources, the considerations are different. You first need to gain a basic familiarity with how to read academic journal material and understand the different kinds of papers and study formats. These guides on types of scholarly sources and scientific study designs (and others like them) are useful primers. Again, most of this comes with just getting reps in, but knowing exactly what you’re reading is absolutely critical. For example, the inability to distinguish between medical journal studies and medical journal letters to the editor birthed the myths about MSG being harmful for health that have persisted for over 50 years.
The other essential thing to check for when reading studies in particular is replication. A study whose findings have been replicated many times is generally more trustworthy than a study whose results have not been replicated. Studies can be flawed and people can make mistakes. The more studies agree with one another, the more possibilities for huge mistakes are narrowed. And if, as is sometimes the case, you encounter wildly contradictory information in which a preponderance of studies find seemingly opposite results, put a pot of coffee on, because it means you have to dig a lot deeper. (This is where contacting an expert can help out most.)
Sometimes, especially when the topic of your research is historical, you may find that the best sources are books. Don’t shy away from them. Keep in mind that while reading a book in its entirety is great, it’s often not necessary. Learning the art of skimming and word-searching can plumb relevant information out of hefty tomes in a tiny fraction of the time.
Digital Jaywalking
Academic researchers have resources the average person doesn’t, including access to paywalled journals and publications and a variety of expensive, subscription-based research services such as JSTOR and LexisNexis. Some of what’s in these databases is extremely difficult to find elsewhere, such as old media and newspaper archives. The lion’s share, however, is accessible to anyone willing to use paywall circumvention techniques, shadow libraries, or peer-to-peer services. Putting aside the dubious morality of charging a fee for information often gathered with publicly funded research, getting around a paywall to access something that would otherwise require a payment is, of course, technically illegal. As such, I cannot advise doing so. I can only point out that jaywalking is also technically illegal, and that everyone does it in at least some circumstances. That’s not an endorsement, it’s a simple empirical observation.
Freely available research can be found through search engines like Google Scholar and platforms like PubMed. Freely available books and archival material can be found through sites like archive.org and Gutenberg. This is all public domain. Archiving websites such as archive.org’s Wayback Machine and archive.is are tools that can not only retrieve deleted web pages or find previous versions, they can also be used to circumvent some paywalls. Other sites include paywall-bypassers like Sci-Hub (for studies and scholarly sources) and Freedium (Medium posts2) and shadow archives like LibGen (books) and Anna’s Archive (books, articles, research, etc.).
Stats, Trends, Polls, and Surveys
Online searches can turn up a lot of data, but these are a few sources I’ve found are consistently reliable, though they’re far from the only ones. For opinion polling: Pew Research, Gallup, and YouGov. For historical or media trends: Google NGram and New York Times Archives. Macro-level data: Our World in Data and Statista. And when it comes to data about economies, crime, demographics, agriculture, etc., government websites and databases are often among the best sources — look for the “.gov” in the URL as a quick rule when searching. International institutions such as the UN, World Bank, and IMF are generally good data sources as well.
Mindset
The most important aspect of research is going in with the right frame of mind. It starts with giving shit about what’s actually true, not just what you’d like to be true. Ideally, it’s best to go into research without any opinions, fueled by a simple desire to follow the evidence where it leads. But we’re humans, with opinions, biases, and priors. When researching, we should do everything in our power to push those aside and simply follow the facts. Personally, I’ve found that some of my best writing was born out of research in which I came to realize that what I thought was true going in turned out to be false. Don’t think of it as being proven wrong or foolish. Think of it as an opportunity to grow.
Enjoying the process also helps immensely. I’ve never been a big fan of puzzles — literal or figurative. I don’t do crosswords, sudoku, Wordle, or Rubik’s cubes. I rarely watch or read mysteries. In the abstract, I have little patience for or interest in such things. But finding out what’s true — digging through oceans of information to understand an issue, point of view, or under-explored facet of society, or to find patterns, trace histories, fact-check a narrative, or simply slake my curiosity — that's the kind of real-world puzzle I’ve grown to find fascinating, even thrilling. That may sound bizarre. I’d have thought so 10 years ago. Research is something most people only ever have to do: for school, for work, for some unavoidable financial or medical decision. It’s only natural to associate it with mind-numbing drudgery. But intrinsically motivated research is a joy I’m glad to have discovered. The right mindset and framing really change everything, including not only whether you do something in the first place, but how well you do it.
See also: “How To Read More: Unlock Your Inner Bookworm”
Subscribe now and never miss a new post. You can also support the work on Patreon. Please consider sharing this article on your social networks, and hit the like button so more people can discover it. You can reach me at @AmericnDreaming on Twitter, @jamie-paul.bsky.social on Bluesky, or at AmericanDreaming08@Gmail.com.
A process I call “gastro-epistemology.”
Honestly, it may be more of a crime to pay for Medium content.
In your Presearch section, you mention alternative search engines not being to your liking. Would you like to contribute your vision to enhance Presearch.com to your liking?