Glossary of Terms

Resource Kit Indices

Just Added

Home > Getting Started > Article >

Speculative and Investigative Searching

by Carl Heine, Ph.D.

I had an epiphany of sorts while attending NECC 2007 in Atlanta. For some time, I had been wrestling with the process of searching, in particular, something I had been calling 'secondary searching.' Whenever I engaged in evaluating a Web page, it seemed to me that this was a different type of search than finding the information I was evaluating.

Then a better word came to mind: investigation. This type of searching involved different techniques than the usual look-for-information-in-Google type of search. There was some of that but there was also truncating a URL to discover ownership, using special operators to reveal hidden references and, among others, searching "Deep Web" databases to track down elusive authors and publishers.

If this is investigation--a word that really seems to describe the process of information forensics--what is the other type of searching? For a while I played around with 'exploration' but that wasn't quite right. Then one day Dennis O'Connor and I were talking about this and he used the word 'speculation' in a sentence. That was it! The other type of searching all of us do most of the time is speculative searching. This is the common search experience: we are not absolutely sure what words to search with nor are we sure where they will be most effective.

There are two fundamental purposes for searching online: to find out something and to verify that it is true. We have to speculate when it involves questions that involve the 1 in 5 Rule. The chances are stacked against us. On average, 4 out of 5 keywords words we use will not be effective. We have to do our best to guess what words may be used in the context of the information we need. Moreover, we don't know for sure if Google has indexed the information or whether only Yahoo! can retrieve it (this happened to me recently in a particularly frustrating search). We have to take chances whenever we search for answers to any of the following questions:

• How many buffalo are there in North America today?

• What is the highest freshwater lake in Tibet?

• What is the top speed of earth's fastest animal?

On the other hand, once we find information that answers our question, the nature of the search changes. Now we have one or more solid clues and can use these to pursue additional information and/or evaluate the credibility of what we found. Investigative searching tends to involve different questions than speculation:

• Who is the author of the Harry Potter Sorting Hat Personality Test?

• Who links to New Zealand Golf Cross?

• Who published the Elect Susie Flynn for President Campaign?

This table summarizes the main differences as I now see them:

More about these two types of searching

Implications

Understanding the difference between the types of searching impacts how the Digital Information Fluency model is used.

DIF Model: Digital Information Fluency

This model is, at best, an idealized representation of how to think about conducting a search. Are there other ways to search? Of course. But fluent searchers tend to conduct organized searches. This model attempts to provide a structure that enables a searcher to follow an effective sequence and know where he or she stands in terms of the whole process.

At a minimum, the circle is navigated twice. The first time around constitutes the speculative part of the search, during which time important concepts and keywords are identified, a place to search is selected and a combination of querying and browsing occurs until information is located. Then comes the all-important question: How good is this information? If it appears not to answer the original question, a "revision decision" is needed: re-thinking what to look for, where to look and how to get there. It is possible to go around the circle many times during the speculative phase, each time trying different keywords and/or places to look.

When information is located that appears to answer the original question, then the investigative phase starts, not by exiting to the middle, but going back to the top. During this phase, the task is to find information that verfies the credibility of information already found. This requires a different type of searching, building on known clues to locate information about the expertise of an author, the reputation of a publisher, the date of publication, the accuracy of a claim, and so on. Each of these discoveries may involve a circumnavigation of the model. Once all the necessary criteria have been found and it is determined that the information passes the test of credibility, then it is finally possible to create a citation, the last step.

In practice, these trips around the circle eat up more energy than time. Sometimes the first speculative search is a winner; then it's just a matter of finding the name of the author or other citation information. Other times, the information remains elusive and it may take a dozen experimental queries before homing in. When that happens, the biggest toll is on one's patience. Investigative searching can go quickly as well, unless missing information was never published. When that happens, knowing when to stop is paramount.

It turns out to be easier to operationalize speculative searching than its investigative counterpart. The following flowchart (from Question and Answer: Searching vs. Browsing in this Kit) may be used with students to encourage strategic thinking when starting a speculative search:

Speculative Searching

A flowchart illustrating decisions associated with Investigative Search will be included in a future update of this article. Click on an icon to learn more about this type of searching: