Good strategy for finding statistics to support writing?
July 16, 2022 4:28 PM   Subscribe

What are some effective and practical ways to find statistics to use as citations in business and tech articles?

I write articles on business and tech industry topics at an unpaid amateur level. I often have intuitions based on experience, e.g. "most tech workers are working in companies with over a hundred employees". And I want to make it an easy habit to support these claims with citations from other people's research. Or even better: prove my hunch wrong when the facts say different, and fix my claims to match reality.

I know how to Google and read through results. But often, I get references like this one where their conclusions are once or twice removed from actual sources, and their citation links go to 404s, paywalls, blank pages, and other dead ends. So it's a goal to avoid sorting through "middleman" citations to arrive at people who actually conducted studies.

I feel like I could easily begin with a simple question like "how many companies in the United States are start-ups?" and get lost in 8 hours of research, finding many close-but-not-same questions, outdated information, broken references, and untrustworthy sources. And then maybe I'd arrive at a decent citation at the end of a day.

If I aim to do a lot of statistic citing in my articles, what are some tools and techniques that will make it go better? Also, I am willing to spend something like $100/month on tools/memberships/consulting. (Not soliciting offers - just saying I have a small budget to put towards solutions.)
posted by ErikH2000 to Writing & Language (3 answers total) 1 user marked this as a favorite
 
In my part of the world, public libraries sometimes have subscriptions to interesting things such as ibisworld market research reports & data - if you have a free public library card, you may be able to access things like this. Sometimes universities also allow alumni access to data resources through the university's subscriptions.
posted by are-coral-made at 4:58 PM on July 16 [2 favorites]


When I need general statistics about a country, I'll use the CIA World Factbook. For more detailed stuff, like numbers on race, religion, language, etc., Wikipedia is often helpful.

There are also often government sources which are excellent - for Canada, Statistics Canada has tons of information on labour markets, demography, etc. Most countries probably have an equivalent.

Finally, if I need information which is quite specific, I would use Google Scholar to find an academic source. Most articles will be paywalled, but you can usually see the abstract and that often has some key data in it.
posted by jb at 7:44 PM on July 16 [2 favorites]


Best answer: Part of it comes with experience...after you cover something long enough you start to know which nonprofit/think tank/industry org/academic expert/etc is likely to have the most up to date data. I typically reach out directly to ask for a stat when I'm on a deadline, since what's online may not be current (you may not get a response quickly if you're not working for a large publication, though). It's pretty common that stats are outdated. Even new academic papers sometimes have stats that are less recent than other sources, presumably since publishing studies takes a long time. I also often see huge differences in stats because of definition- in the example of the number of startups in the U.S., I think people could reasonably disagree what a "startup" even means. The Census Bureau is tracking new business formations by month now. But many companies that are, say, four or five years old still consider themselves startups. Crunchbase has a different list. (I don't focus on startups, but that's probably where I'd turn in that example, or maybe the Kauffman Foundation.) I typically have short deadlines, so I try to explain the context/limitations of each stat and let go of trying to find perfection.
posted by pinochiette at 6:37 AM on July 17


« Older Best Bike gadgets, accessories, and add-ons?   |   Meth is supposed to be odorless Newer »

You are not logged in, either login or create an account to post comments