Searching for Online Resources: Strategies and Search Engines

I tried to put myself in the position of a college instructor who is designing a new course and wants to see what online instructional resources might be available to use in supplementing the course. My guess was that an instructor would most likely begin with a Google search, since Google is currently the most popular search engine (and is built into the top-level search choices of many browser home pages).

A general Google search for "learning resources" on the web revealed a few interesting resources, but failed to show any of the major resource repositories among the first several display pages (at 10 sites per page). The searcher would need to be very persistent and look all the way to the 8th page of the listings where MERLOT finally appeared in 80th place; MIT's OpenCourseWare site did not appear at all among the first 10 pages or 100 sites, nor did SMETE, Connexions, Alexandria, Deliberations, GEM, LTSN, or the Scout Archives. This Google search pulled 509,000 citations.

A Google search for "learning objects" on the web did luckily show one relevant site more quickly, with the Center for International Education's useful web page on Learning Objects (http://www.uwm.edu/Dept/CIE/AOP/learningobjects.html) showing up as the 4th item on page 1 of the search. This site would lead the searcher to MERLOT and other general resource repositories if she or he were inclined to look at it. However, none of the major general resource repositories showed up directly within the first 10 pages. Most of the citings about "learning objects" were theoretical discussions of metadata, definitions of learning objects, and other technical articles. This Google search pulled a total of 69,700 listings.

At this point I decided to do a narrow search within a specific discipline, Psychology. (I chose Psychology because my Ph.D. work was in the field of Experimental Psychology and I've taught a number of different courses in the field from Introductory Psychology to more specialized upper division and graduate courses.) A Google search for "psychology learning resources" brought up only 5 pages of 47 citations (with many duplications), none of them led to MERLOT, OCW, or any of the primary learning repository sites, nor to a specific discipline-based repository for psychology. A search for "psychology learning objects" failed to identify anything, "your search did not match any documents."

A general Google search for "psychology" pulled up 7,610,000 listings! Many of these listings were worthwhile, including the Encyclopedia of Psychology (http://www.psychology.org/) appearing on p. 1 in position 2 and the Psychology Virtual Library (http://www.clas.ufl.edu/users/gthursby/psi/) that appeared on p. 1 in position 6. On page 2 the useful Psychology Online Resource Central appeared (http://www.psych-central.com/) with many useful links to more online resources in the field.

It's clear that a searcher who is uninformed about learning resources/objects is likely to stay uninformed by doing general Google searches for "learning resources" or "learning objects" unless she or he is very lucky. The most useful strategy for a novice searcher to take would be to do general searches within her/his field rather than general searches for "learning resources/objects" because these searches do not easily lead to the identification of the primary learning resource repositories such as MERLOT or OCW. (Of course, a well-informed searcher would begin with MERLOT or OCW and then supplement those searches with Google browsing or browsing using other search engines.)

I repeated the general and specific searches by using a different, science-oriented search engine, Scirus (http://www.scirus.com/about/). The Scirus search for "learning resources" did detect the Dept. of Education Gateway to Educational Materials (GEM) site (http://www.thegateway.org/) on the 3rd page as item 25 among the total of 175,690 items that were identified. The same search also elicited many discipline-specific resource sites on every page. However, this general search did not identify MERLOT, OCW, SMETE, or other key repository sites within the first 10 pages (or 100 citations).

The Scirus search for "learning objects" produced 18,869 citations. Unfortunately, for our hypothetical novice searcher, almost all of the citations consisted of theoretical discussions about learning objects, the construction of learning objects, and metadata standards; no direct link to a learning object repository appeared within the first 10 pages (the first 100 citations).

A narrowed Scirus search for "psychology learning resources" yielded a nothing but the message, "Sorry, your search has not produced any results." Similarly, a narrowed search for "psychology learning objects" failed to provide any citations.

Going to a discipline-wide "psychology" search within Scirus provided 2,288,861 citations! Clearly the searcher would need to narrow the search to specific areas of psychology and then try to filter the citations to locate those that contained online learning resources. E.g., searching with the term "cognitive psychology" pulled in 88,394 total citations. By page 3 item 25 appears with a citation of PsyQuest materials that could be applied in a course or section about Cognitive Pyschology (http://cc6.cumber.edu/psych/home.htm). Item 26, the CogPsychTutor, also provides some potentially usable materials (http://teach.psy.uga.edu/CogPsychTutor/default.htm). Although this route may yield some usable resources, it is a slow way to find them and the searcher may never discover some of the valuable resources that are available at the major online repositories. An uninformed searcher is better off doing general searches within her/his field and using that as a pathway for locating reusable learning objects, but that is a slow, laborious way to find information. Most instructors will be more likely to rely upon resources supplied by textbook publishers rather than engage in time-consuming searches for online resources.

It's an old quandary that we don't know best how to get somewhere until after we've already been there. The novice searcher is in the familiar, uncomfortable position of not being able to find what she/he is looking for until knowing where to look and what exactly to look for. It's like walking through the woods looking for a trail; once you find it, if you do come across it and can recognize it, everything is easy--you can travel where you want to go, but until then everything is bafflingly difficult. It's hard to find what you want until you know where to look and know how it's labeled. A searcher new to the mountain of learning resources and learning objects would not necessarily pick up MERLOT in a general search or be able to identify it as a primary resource site; a better informed searcher would start with MERLOT, OCW, GEM, Scout, and some of the other primary sites. After using the primary resource sites to find learning resources it could still be useful to try some broad general searches, just to see what serendipitously emerges, but reliance on serendipity should not be the first or only step in an intellectual exploration of online resources.

The conclusion is straightforward: to facilitate the use of online learning resources it is important that portal sites be created to map the way to important repository sites and provide guidelines for their effective use by instructors. (It's also clear that the semantic web developments need to continue so that web searches can become more productively selective; the web needs an informational GPS.)