In last week’s article I wrote about the importance of verifying information. This week, I’ll examine a strategy on how to go about doing so.

When it comes to looking up information, you’ve likely heard the oft-repeated adage, “Google is your friend.” Well, it is true that Google can be very helpful, but there’s something you should know about your “friend.”

Google is like your typical research scientist: by day he’s clean, caffeinated, and brilliant, but by night he’s a raging drunk. Perhaps you’re friends with him, but it’s probably healthiest to just call him an acquaintance on cordial terms. At parties he sits out on the porch, slumped in a plastic chair in a boozy stupor, waiting for someone inobservant enough to sit next to him and strike up a conversation. Provide him with some ambiguous keywords (e.g., “hey man, how’s it goin’?”) and he’ll vomit all sorts of unwanted, incoherent, and occasionally X-rated gibberish. However, if you were to provide him with precise, targeted keywords (e.g., “hey man, I heard you study the biochemical underpinnings of pattern baldness. Would you mind explaining to me the current understanding of the role that the androgens play in male androgenetic alopecia?”) and… well, he is still drunk. You might get some on-topic material, but probably nothing of substance, and he’s guaranteed to leave something important out. Pretty soon he’ll wind down, lurch to his feet and “go check on a gel” in the bushes behind the house. BUT if you wait until he sobers up a bit, you could get a very clear, detailed, and up-to-date summary of what you want to know.

My point is you need keywords that are specific and targeted and you need to know where to submit these keywords. Google (www.google.com) is our scientist friend by night. If you’re just looking for a simple recipe for a vodka martini, he could be a great reference, but for reliable information on complex topics, we need his sober alter ego: Google Scholar.

Google Scholar (scholar.google.com) is my go-to source for detailed information I actually care about. It pulls up peer-reviewed journals and literature reviews, rather than your typical web pages. The search function is free, and very, very good (better than what you can find in most science journal databases).

Note that, unfortunately, if you don’t have access to a school or university’s journal subscriptions you may be unable to access many of the articles that the search turns up without paying a fee; however, most searches will usually turn up a few useful, freely accessible articles. Additionally, the authors themselves will sometimes post full-text versions of their articles on their own websites.

I first started using Google Scholar in grad school, searching for papers relevant to my research; however, I quickly found that this can be a great way to find reasonably-sussed-out information on pretty much any topic. For example, as a new parent, you can imagine the sheer volume of questions that I come up with. You also can image the garbage that turns up if you dare to include the keyword “infant” on a standard Google search – all sorts of heavily skewed articles written by people trying to get you to buy their product. In comparison to the tumult of a standard Google search, rummaging through the primary literature and lit reviews can be a refreshing jaunt.

So, now we know where to search, but how do we come up with appropriate keywords? Choosing the right keywords can be challenging – especially when you know very little about the topic that you’re researching. The trick is to start with a general search specifically to mine for keywords, and then use your preliminary findings to do a more targeted search.

For example, I just finished making a lecture for one of the classes that I teach that peripherally addresses male pattern baldness. The textbook that we use implies that there is a single gene involved in pattern baldness. Right away, this didn’t sound right – even anecdotally, I could think of friends and acquaintances that had baldness in their families that was distributed in patterns that did not seem to adhere to this relatively simple mode of inheritance. I suspected that there was more to it than the textbook suggested. I checked a couple other textbooks I had on hand, but lacking any specialist texts, I was unable to find any further information.

So, I turned to the internet. But I knew next to nothing about baldness, save for the suspected faulty information in the textbook. What do I search for? A general search on Google reveals all sorts of quack baldness remedies. Not helpful so far. Wikipedia, of course, floats to the surface of the list of results like a dead fish in oxygen-depleted water – kind of gross, but, you know, sometimes worth poking with a stick. This is actually one place where Wikipedia can sometimes be of limited use.

First of all, it’s important to keep in mind that ANYONE can author a Wikipedia article, and ANYONE can modify an article. While the folks at Wikipedia I’m sure do everything they can to keep the information quality as high as possible, there are no guarantees. But there are a couple things that Wikipedia IS good for:

1. Finding keywords
2. Finding lead-in sources with more information
3. Sometimes catching a whiff of any controversies or biases that may embroil your topic

A quick scan of the relevant Wikipedia article mentions that male pattern baldness may also be called “androgenetic alopecia” as well as a couple other specific terms; already we’re off to a great start – we have a specific name. Sometimes, if I happened across a particularly interesting or provocative comment, I’ll look up its source and follow up on it, but in this case I don’t know enough about the topic to even recognize anything as particularly unusual. Additionally, if I suspected that some of the sources I might encounter later in my search would be biased, I might find mention of specific controversies here (for example, if I were researching pattern baldness treatments instead of causes, I would imagine that there would be a lot of bogus claims to be on the lookout for).

At any rate, I’ve now got a couple keywords, which is enough to get me started at Google Scholar. Once there, I select my time range in the sidebar. For a hot topic such as this I’ll want to keep the date range relatively current; 3-4 years back is probably adequate. I don’t want any studies that came out in the fifties – I’m assuming that pattern baldness is an active area of research, and that older studies will quickly become outdated. If I were researching a somewhat more obscure topic, I would push back the date range until I found something useful.

Next I’ll enter in some of my keywords, including the word “review.” My goal here is to track down a literature review – this is where a scientist scours the literature and summarizes the state of affairs in their area of research. A good lit review usually cites many of the formative studies in the field, so you have a list of the topic’s “greatest hits” to follow up on. A well-written literature review can be pure gold – some authors do a great job at distilling a complex subject into its most important constituents.

So, now we have a list of search results. At this point, I’d like to point out a really helpful feature of the search engine: in the lower left of each search result it reads, “Cited by [number].” This tells you how many other peer-reviewed journals have cited that particular paper.

More citations doesn’t necessarily mean that the information in the article is necessarily correct, but it does mean that people are talking about it – generally (but not always) a positive sign (though it could also mean that it’s so far out in left field that people keep talking about how crazy it is). Few citations, on the other hand, can mean: 1) the article just came out, and nobody has had a chance to cite it; 2) the article is in an obscure or poorly-studied field, and while the information may be great, nobody really cares; or 3) the information is total garbage. Number of citations is definitely something worth keeping in mind when sifting through the results.

Anyway, to make a long story short, I found a well-written review that supported my suspicions that male pattern baldness is likely to be related to a number of different genes; it also provided me with some other interesting tidbits, and of course references. I followed up on some of these references, ran some more searches with new keywords gleaned from the review, and found further confirmation of the data in the review. Of particular interest to me was that, while there is by no means total agreement in the mechanisms of baldness, no current sources I encountered suggested a single gene to be entirely responsible for it. I didn’t need to investigate the subject in too much detail because it wasn’t necessary for my class, but interestingly, a couple articles into my research I found the likely source for the textbook’s misinformation: a poorly designed research project from 1916; widely cited, but generally discredited.

So there you have it: the basics of how to verify your information. Remember, you can apply this process to pretty much ANY QUESTION YOU HAVE. One final note: this will take some practice! Some of the information you find will likely be over your head – there’s a lot of specialized terminology and high-brow concepts out there – but don’t get discouraged. With a little practice you’ll find that you can extract the main conclusions without too much hassle. Do you have an alternative to Google Scholar that you prefer? Or any other useful tips in verifying information? Let’s hear about them in the comments below!