Yesterday was back to school day, and I had the pleasure of spending it with fabulous librarian Anna Koval of the Big House Library. We were discussing student research projects, and what—given a limited time—students need to learn first about using search engines.
I believe that the advanced search engine functions
(site:, filetype:, and the like) are amazing tools for the thoughtful searcher,
but the very first skill one has to learn is the thoughtfulness itself. Since search engines essentially match the
words you enter, the number one rule for good searching is simply this: really
think about what you put in, and better quality stuff will come out.
This is especially difficult when you are looking
for “invisible” concepts—that is, ideas that are often expressed without the
words that we use to define them. Think “statistics.” We all know what
statistics are—we “know them when we see them”—but they are often recorded
without actually using the word itself. Other times, the kinds of statistics we
want are expressed in so many different ways it is hard to know which ones to
ask our search engine for. If we are simply looking for percentage of US
citizens who are homeless, just that idea can be expressed as percentage, percent, or %. Search engines like Google will look
for the former two terms, but cannot even see the latter. In the same way, if
one of Anna’s seniors wanted to find forecasts
for childhood obesity, they would also have to contend with predictions and projections.
So, this is where visualization really comes into
play. Say I want to know various public
policy goals regarding childhood obesity. In past years, I would have used
visualization to brainstorm a list of search terms—imaging all the different
ways I had seen an idea expressed: [“childhood obesity” priorities OR goals OR
targets OR seeks OR returning]. Now, however, I am working smarter. I really
think about the sentences, and realize there is one commonality in the writing
I see about the topic when I search Google:
"..implement our six policy priorities for reversing the childhood obesity epidemic by 2015...."
"...The goal, as set out in a report from the White House Task Force on Childhood Obesity, is to reduce childhood obesity from 20% to 5% by 2030...."
"... The report targets a real cut of 2.5% in obesity rates by 2015 and another ..."
The common form of expression in every single sentence
on the topic is actually the phrase “by 20xx.”
Generally speaking, the years projected tend to run from 2015 to around 2050,
meaning that the concept can be easily retrieved making use of Google’s
excellent number range search (..) with the search: [“childhood obesity” “by
2015..2050”], with which I ask for pages with the phrase “childhood obesity,”
and the phrase “by 2015,” or “by 2020,” or “by 2025,” and so on, up to “by 2050.”
So Anna and I started discussing how we could introduce
this kind of visualization to her students. Anna introduced me to AwesomeHighlighter.com, a tool
for virtually highlighting online articles, and we practiced one of my favorite
skills—reading a Washington Post article on childhood obesity not just for
information, but also for search terms that can be used to build enhanced
searches.
There are three categories of search terms we
found:
1) 1) basic terminology we would need to search
for more information, such as obesity
epidemic;
2)
2) sources of more authoritative information,
such as the White House Task Force on
Childhood Obesity; and,
3)
3)associated search terms, such as by 20xx, which appeared four different
times in the article with related statistics.
Teaching how to read for terminology and authorities
is relatively straightforward. Reading for associated search terms is harder,
but should drive home the concept that visualization can suggest creative and
productive queries. So it is a tricky, but worthwhile, task.
My goal is to experiment over the next few months,
and start to build a list of associated search terms for searching news
sources. I’ll be sure to share. Can you think of any more examples to help me
get started?
