Mining Industry Data: A Checklist Approach
Ojala, Marydee, Online Searcher
Two years ago, in the January/February 2011 issue of ONLINE, I wrote about checklists for private company information. It was well-received, so I thought I'd try my hand at creating one for industry research. This is much more challenging for one important reason. While there is a paucity of information sources available for privately held companies, the opposite holds true when researching an industry. Industry information is plentiful. The trick is to narrow the abundance of information down to something usable.
I've used the mining industry as my example to amplify the checklist points because it's a worldwide industry, with many subsections and the object of numerous news stories.
In The Checklist Manifesto: How to Get Things Right (New York: Henry Holt and Co., LLC, 2009), surgeon Atul Gawande makes the case for a checklist approach in his field, medicine. He tells stories about checklists in hospitals and investigates how checklists can improve performance and prevent disasters. Although he doesn't mention business research, information professionals can easily adapt the checklist approach to ensure they cover the salient sources for research projects.
Checklists are written guides that include key steps for complex procedures. Gawande says, "Good checklists ... are efficient, to the point, and easy to use even in the most difficult situations. They do not try to spell out everything...." Checklists help ensure that you get critical information in a systematic fashion. Gawande stresses that checklists aren't intended to be "comprehensive how-to guides." Instead, he sees checklists as "quick and simple tools aimed to buttress the skills of expert professionals." They also aren't set in stone. When a resource on your checklist changes character, disappears altogether, or prices itself beyond your ability to access it, you should remove this source from your checklist and, possibly, add an alternative.
For industry research, the first step is to determine if there's an applicable industry code. We usually think of NAICS and SIC codes, but don't forget about other applicable codes, such as those relevant in a particular country, tracking international trade, or created for a subscription database such as ABI/INFORM. The high-level NAICS code for "Mining, Quarrying and Oil and Gas Extraction" is 21. This breaks down into multiple subsections so you can drill-down to coal, metal ore, nonmetallic minerals mining, and more.
ABI/INFORM has mining industry as a thesaurus term, but its code is 8500, which ABI/INFORM describes as "extractive industries," with the word "mining" nowhere in sight. Both Factiva and LexisNexis use their own indexing for industries. Factiva, for instance, has imet as the most general code but breaks the industry down into numerous components, using alphanumeric codes such as i211 for metal ore mining.
For some industries, particularly those undergoing rapid change or ones too small to rate their own code, descriptive words are the better approach. Suppose you were looking for emerald mines. The closest NAICS code, 212399, concerns gemstones; emeralds are not mentioned. EBSCO Business Source Premier, however, supplies Emeralds and emerald mining as a thesaurus term.
Geography is another important item. Are you looking for copper mining in Peru, gold in South Africa, diamonds in Canada, or coal in China? Within the U.S., which state or region piques your interest? Some parts of the country are more dominant in a given industry than others. The Census Bureau's Economic Census: Industry Snapshots (www.census .gov/econ/census/snapshots) provides a handy map of industries by state, determined by number of establishments, number of employees, and payroll per employee, down to the six-digit NAICS code level. Colorado, for example, leads for uranium/radium/vanadium mining (code 212291).
DECODING YOUR SEARCH
Determining the relevant codes and thesaurus terms is first on the checklist, but it's not the most important element of an industry research project. …