University of Maryland University College
UMUC Secondary Page Template

Search the Web & Evaluate Web Resources

Contents


Types of Search Tools

The following table outlines a variety of tools used to search for information on the Web. It's a good idea to become familiar with several different directories, search engines, metasearch engines, and specialized tools because no one search tool indexes the entire Web.

Search Tool
Characteristics
Examples
Subject Directories

(also known as "virtual libraries" or "portals")

  • Indexes Web pages selected by editors
  • Organized into hierarchical subject categories
  • Searches a description of Web pages, not the full text
  • May be annotated
Search Engines
  • Uses "spiders" or "knowbots" (computer programs that roam the Web for sites and keep the search engines up to date)
  • Provides keyword searching of words in pages or full text of selected pages
  • May not have subject categories
  • May not search "invisible Web" of information stored in databases
Metasearch Engines
  • Searches several individual search engines simultaneously and compiles results
  • Only catches about 10 percent of the search results of the engines visited
Specialized Search Tools
Google Scholar
  • Searches for scholarly literature, including papers, theses, books, preprints, abstracts and technical reports from broad areas of research
  • Be aware that many fee based articles in Google Scholar are available free in the library's databases
  • For help in using Google Scholar see: Tips for Searching Google Scholar via UMUC

Top


Basic Searching

Searching for ALL or ANY Keywords
Most search engines look for ALL the words in a search. For instance, in a search for dog canine puppy , only sites that contain all words will be retrieved. To complete a search with ANY keyword returned, include an OR between words. For example, dog OR canine OR puppy will return sites that contain one, two, or three of the words.

Capitalization
The capitalization of a letter does not matter in most search engines. Whether you search for Napoleon Bonaparte, napoleon bonaparte, or NaPolEON boNaPArte, your results will be the same.

Common Words
If you enter short, common words, such as a, and, the, who, how, and in, a search engine or directory will eliminate those words from your search. If you need to include these words in a search, such as for whom the bell tolls or in the daylight , be sure to put them inside quotation marks for phrase searching - "for whom the bell tolls" or "in the daylight".

Top


Search Engine Features

The table below describes common features available in many Web search tools. Please note, all of these tools are not available in all search engines or directories. Check the help and advanced search tips pages of the search tool you are using to see if these features are available.

Feature
Symbol
Example
Action
Include a keyword
+
+intellectual +property +patents
Finds Web sites where the term after the + appears in the results.

NOTE: Can be used to override the default setting where ANY of the search terms are retrieved.
Exclude a keyword
-

+intellectual +property -patents

Finds Web sites with intellectual and property, but not patents.
Search for a phrase
" "
"four score and seven years ago "
Finds Web sites which contain the exact phrase.
Search within the title of a Web site
title:
title:mortgage
Finds Web sites with mortgage in the title tag.

Search within a site domain

site:

cherry site:www.pepsi.com

Searches within the site www.pepsi.com for the keyword cherry
Search using truncation
*
finan* planning
Finds Web sites with finance, finances, financial, etc.
Use Boolean operator AND
AND
montana AND camping
Finds Web sites with both terms.

Use Boolean operator OR

OR

montana OR camping
Finds Web sites with either term.
Use Boolean operator NOT

NOT

montana NOT camping
Finds Web sites with montana, but not camping.

Search using nesting
( )
(college OR university) "bookstore"
Finds Web sites with college OR university, and bookstore.

Top


Ranking the Results

Search engines use algorithms to sort the results and rank them according to relevancy. Presumably, the most relevant sites appear at the top of the list.

But how does the computer decide which sites are "relevant"? Most search tools do not reveal their ranking formulas--protecting their competitive edge--but they do share some elements of the criteria they use to rank the results.

Some elements considered in ranking the results are

  • Number of terms matched
    Web pages containing ALL of the search terms will be ranked higher than Web pages with only some of the search terms.

  • Proximity of terms to each other
    Web pages where the terms appear closer to each other (i.e., in the same section, paragraph, or sentence) will be ranked higher than Web pages with the terms separated.

  • Location of terms within the document
    Web pages are created using HTML (HyperText Markup Language) which uses a set of tags to indicate how the content of the page should appear. The information in some of those tags is used by some search engines to determine relevancy. For example, some search tools give higher ranking to Web pages where the search terms appear in the title, head, or meta tags of a particular Web page.

  • Frequency of terms in the document
    More occurrences of the search terms in the Web page give a higher ranking. However, most major search engines recognize "spamming" techniques and discount them in the ranking. Spamming is when the Web page designer loads terms in the page or repeats the same few terms over and over to improve the ranking. Most search engines can detect this and either lower the ranking of the site or ignore it entirely.

  • Number of links to the page
    The more sites that link to a given page, the higher that page appears in the rankings. This technique is how the search engine Google ranks the results.

Top


Evaluating Web Resources

The World Wide Web's accessibility and ease of use has led to an exponential increase in the amount of information available on the Internet. As a research tool, however, the Web lacks the quality assurance that editors provide with print publications or that librarians provide when collecting materials for their library. Therefore, given that some Web sites are out-of date, biased, or inaccurate, it is crucial that people doing research in this medium learn how to critically evaluate the material they find.

The following checklist, divided into five separate criteria for evaluation, presents questions that should be asked to determine if a Web resource is of high quality or not. Go through each of the criterion and answer the questions. The greater number of questions answered "yes," the more likely that the source is of high quality.

Criterion #1: Authority

  1. Is it clear who is sponsoring the creation and maintenance of the page?

  2. Is there information available describing the purpose of the sponsoring organization?

  3. Is there a way of verifying the legitimacy of the page's sponsor? For instance, is a phone number or address available to contact for more information?

  4. Is it clear who developed and wrote the material? Are his/her qualifications for writing on this topic clearly stated? Is there contact information for the author of the material?
Criterion #2: Accuracy
  1. Are the sources for factual information given so they can be verified?

  2. Is it clear who has the responsibility for the accuracy of the information presented?

  3. If statistical data is presented in graphs or charts, are they labeled clearly?

  4. Are there errors you can substantiate in the data presented?
Criterion #3: Objectivity
  1. Is the page and the information included provided as a public service?

  2. Is it free of advertising?

  3. If there is advertising on the page, is it clearly separated from the informational content?

  4. Are there any other signs of bias?
Criterion #4: Currency
  1. Are there dates on the page to indicate the following:

    • a. When the page was written?
      b. When the page was first placed on line?
      c. When the page was last revised or edited?

  2. Are there any other indications that the material is updated frequently to ensure currency of the data?

  3. If the information is published in print in different editions, is it clear what edition the page is from?

  4. Are the links on the page up-to-date?
Criterion #5: Coverage
  1. Is there an indication that the page has been completed and is not still under construction?

  2. If there is a print equivalent to the Web page, is there clear indication of whether the entire work or only a portion of it is available on the Web?

  3. If the material is from a work that is out of copyright (as is often the case with a dictionary or thesaurus), has there been an effort to update the material to make it more current?

  4. Is there any other evidence of omissions?

  5. Does it cover the subject adequately?



The above checklist was based on the following WWW page: 

Alexander, J. & Tate, M. (1996, August 8). Teaching Critical Evaluation Skills for World Wide Web Resources. Retrieved on March 13, 2006, from http://www3.widener.edu/Academics/Libraries/Wolfgram_Memorial_Library/
Evaluate_Web_Pages/659/

Top


Using the Web for Research:
What's good on the Web?

Some types of Web sites can be particularly useful for academic, scholarly, or professional research. They are:

  • Company Web sites
    Information on products, services, staff, and the general organizational structure of the company.

  • Government Web sites
    Information on almost anything! U.S. government produces tremendous amounts of information in a variety of areas.

  • Organization Web sites
    Information on the issues an organization advocates and the general organizational structure of the organization.

  • Statistical Web sites
    Statistical information and data sets from government and academic pages.

Visit the links above to learn more about what kinds of information you can find from these Web resources.

Top