~~~~~~~~~~~~~~~~~
in our search for periodical articles
Where we’ve been —
PERIODICAL STACKS
Last week you needed to find an article from a magazine or a journal on a particular topic. So, you went to the library and wandered around until you saw a journal that might be related to our topic. Then you flipped through it and maybe found an article of interest.
As you know from experience, this is a very inefficient way to look for articles, but some students really do it this way! You only needed one article on a very broad topic, e.g. “earthquakes.” Imagine if you needed 7 articles, not simply on earthquakes, but on engineering methods that help buildings remain standing during an earthquake! Finding these articles by roaming the stacks could be a very long-term project.
PERIODICAL INDEXES (IN PRINT FORMAT)
After your random search you quickly learned about periodical indexes, such as the Reader’s Guide to Periodical Literature, which allow us to look up a subject and find a list of articles from different periodicals on that topic.
This approach is MUCH better than roaming the stacks, but it is still time consuming. For example:
- Each physical book, e.g. each volume of Reader’s Guide, only contains indexing for one year. This means that if you look in the 1990 volume for your topic, you will find articles from 1990. Suppose you also want articles from 1995 and 1996. For each year, you would need to repeat your search in a different volume of the index.
- As you find articles of interest, you need to write down all the information given about the article, making sure you copy it correctly and don’t leave anything out. In some cases, you will need to look in another spot in the index to convert an abbreviation for the periodical title to the full title of the periodical.
- Finally, you need to rely on the subjects provided by the index, which, like Library of Congress Subject Headings, don’t always make sense (e.g. the LCSH for “babies” is “infants”). The subject listing in the index can also make it difficult to search for a combination of topics, e.g. exercise and cholesterol. Often you have to check both subjects and scan the articles under each to find relevant articles.
In our fast-paced, 21st century life, there must be a better
way…
$%^&$%&*%^*^
Fortunately, most periodical indexes are now available in electronic format, either on the Web or on CD-ROM, which makes searching for articles much easier and faster.
As we look more closely at Reference Databases over the next couple of weeks, remember that Reference Databases are the same thing as periodical indexes in book format - they have just been given an electronic, searchable interface. In many cases the online versions of an index correspond exactly to a book-like index that is in print format. For example, we have the MLA Bibliography on the Web as a reference database, and we also have a paper index titled MLA Bibliography in the Reference stacks.
Computerized reference databases have three main advantages over traditional print indexes:
1. Many years of articles can be searched at once. For example, I can do a search on airplane hijackings and find articles from 1985 to 1995 in one search.
2. Two or more search terms can be combined. This results in a more sophisticated search then just looking up a subject in a print index. For example, in the Reader’s Guide we looked for articles on one topic or subject heading. What if we wanted articles with more focused topic, such as the effects of exercise on preventing cancer? In a reference database, we could type something like “exercise AND cancer” and come up with a list of related articles.
3. Finally, reference databases allow searching by a greater number of access points. An access point is a way to access information in a catalog or index. For example, in the HELIN catalog, we can search by author, title, subject, call number, etc. All of these are access points. In the Reader’s Guide one can search by subject or author only. In a reference database, you can search by author, title of article, subject, title of periodical, date published, and so on. This allows you to run more powerful and precise searches.
There are many general and subject specific reference databases available to us at URI. From the URI Libraries homepage find and click on the link for "Reference Databases." You'll come to a page that lists major disciplines of knowledge (Arts, Business, Health Sciences, etc.) Each link will connect to a list of electronic databases, most of which are subject specific and fall in a subject category, and some are general and cover all topics. Some index only scholarly journals, while others cover all types of periodicals.
Go to "All Databases by Title." Note how many databases URI has.
Also, like indexes such as the Reader’s Guide,
URI doesn't make these databases - we buy them from companies that provide access to them
for a fee. Many are quite expensive, over $10,000 per year. Most are less, about $5-6,000
per year.
RRRRRRRRRRRRRRRRR
Academic Search
Premier
(from URI Libraries)
You'll see that Academic Search Premier:
- Indexes different types of periodicals – popular/trade/scholarly
- Covers all different subject areas
- Has a great deal of full text online.
Click on the link and try to do a little self-exploration:
Basic Search (refer to your Search Strategy Work Sheets from previous assignments) Separate different concepts with AND – smoking and depression. You can use OR – (smoking or nicotine) and depression
Limit Your Results
Below the search box there are several limiting functions you
can use before you conduct your search:
- Full Text - only articles that include full text will be retrieved (not recommended)
- Peer-Reviewed - only articles from scholarly (or peer-reviewed, refereed) journals will be retrieved.
- Journal - restricts your search to the journal that you enter (assuming that Academic Search Premier indexes that journal)
- Number of pages
- Articles with Images
- Date Published
Searching
Do a simple search on your topic - type in your term(s) and
click the "Search" button. Hopefully, you'll come to the Results List, listing
the articles that ASP found using your terms.
Results List
Search results are listed 10 on a page. The total number of
"hits" or records retrieved is shown at the top of the list with links to
additional pages. Each short entry includes the citation to the article as well as other
information about the availability of each article, such as "Full Text,"
"Full Page Image" (Adobe Acrobat PDF file), "Search HELIN" or other
notes.
NOTE: Don’t disregard non-full text over full text – the number of full text articles is based on agreements between publishers and the database vendor, not on the quality of the article. What’s included and what isn’t is arbitrary.
Seeing Results
Click on an article and look at the record. It will consist
of several "fields" - the Title, Subject, Source, Author, Abstract, other
identifying terms, and (maybe) the full text of the article.
The citation includes all the information you need to find the article (and to cite it in your paper's bibliography or your final project for this course.) The citation is made up of several parts - title, publication, date, author, volume. issue, page numbers, standard number and subject terms (some publications include some other fields).
Look at the field entitled "Subject(s)." Is one of your search terms listed there? This field can come in very handy when you are searching for articles on the same topic. The Subject field uses the database's "controlled vocabulary" (discussed below) to identify what the article is about. You can click on any of these terms and bring up another set of articles on that topic.
Can you tell if the articles are from popular or scholarly journals? Review the differences from earlier lessons to help you figure it out. In Academic Search Premier you can limit your search to just scholarly journals right from the search page. Click on the box next to "Peer Reviewed" and conduct the search.
Most articles have an abstract, a few sentences describing what the article is about. If available the full text will follow the abstract (if available).
On the top of the screen you will see buttons for printing, saving and e-mailng the article. The print function turns the record into a printer-friendly version and the e-mail link will allow you to send the entire article.
You can get back to the results list by clicking on the "Result List" button in the menu near the top of the page. You can also Mark the articles you are interested in by clicking the little box next to each article on the list (or in the record itself) and retrieve them by clicking on the "Print," "Save" or "Email" buttons.
Finding Articles Using Citations
Some of the articles in Academic Search Premier will
only have a citation and an abstract. (Also, very few of the other electronic databases
provide full text of articles.) To find the text you'll have to find the journal in the
library. This requires a search for the journal in the HELIN Catalog.
If there is a "Search HELIN" link, click on that and it will search the catalog for you. Otherwise, go to HELIN and "Restrict Your Search" to the "Journals/Serials Collection" and do a Title search for the title of the JOURNAL. (You cannot find individual articles in the HELIN catalog.) There you will find the record for the journal including the Call Number (that's how you will find it on the shelf) and information on which volumes the library has (compare what the library has to the volume and issue number listed for your article).
For more information, consult the URI User Guide, Searching for Periodicals in HELIN .
<><><><><><><><><><><>
ACCESS POINTS & FIELD SEARCHING
An access point is a way of searching using an information tool. Each access point is one part of a database record that provides access to the whole record, for example, title and author are access points when searching the HELIN library catalog.
The most common access points are author, title, subject, and keyword, but reference databases often allow searching by many additional access points.
To help understand access points, imagine that you are building a database of your CD collection. In the database, you want to include every possible piece of useful information about each CD, such as:
- Title of the CD
- Artist
- Artist’s nationality
- Artist’s gender
- Other band this artist has been connected to, for example Eric Clapton would be linked to Cream, Blind Faith, Derek & the Dominoes)
- Producer of the CD
- Other CD’s the producer has worked on
- Tracks on the CD
- Number of tracks
- Name of each track
- Length of each track
- Type of music
- Awards the CD has won, if any
- Record label
- Other musicians on the CD
- Year CD released
Each of these bits of searchable information, or access
points, would be placed in a different field of your database record.
If you included all of this information, you could set up your database so that you could search by any of these fields, or a combination of them, for example you could search for every CD by a particular artist during a specific range of years.
Searching by different fields can also yield different results. For example, searching for Hemingway in the author field of a library catalog will find books by Hemingway. Searching for Hemingway in the subject field will find books about Hemingway.
To see how field searching by different access points works, we’ll use Academic Search Premier as an example.
Just like library catalogs use Library of Congress Subject Headings, reference databases use controlled vocabulary to index articles. That is, they use designated words or phrases, sometimes referred to as descriptors or subject terms, to consistently describe the topics of articles. This makes searching for articles more effective and precise. Just think, if you were making that database of your CDs discussed above, you would want to use the same words to describe types of music so that when you searched you’d get all the relevant records with one search. For example, when you enter your Rap CDs into the database, you will want to decide whether you are going to call them Rap or Hip Hop. If you are not consistent and later you perform a search for all of your Rap CD’s, you are going to miss the ones that you had categorized as Hip Hop.
Each database has its own unique controlled vocabulary.
These terms are often based on the specialized terminology of the subject area covered by
the database, for example: business, economics, zoology, medicine. This vocabulary of
subject-specific databases is often very focused – much more so than the Library of
Congress Subject Headings or the subject headings of general databases. For example, for
the topic allergies:
Database |
Subject Heading |
|
| Library of Congress Subject Heading (Academic Search Premier) | Library catalogs & some databases (ex., ASP) | Allergy |
| ProQuest Direct subject heading | ABI/Inform | Allergies |
| Medical Subject Heading | MEDLINE (PubMED) | Rhinitis, allergic |
Such focus makes it easier to find subject-specific information. Most databases have thesaurusi with which you can check your subject keywords and find out what the database’s controlled vocabulary for your subject is.
STOP WORDS AND PHRASE SEARCHING
So far, we have learned that before searching the library catalog or a reference database, we first identify our main concepts, list keywords that describe each, and separate the different concepts with the word AND, for example smoking AND depression. In general, if we do not separate our concepts with AND, instead typing in words in succession, or even a sentence, the database will not interpret our query correctly. For example, if I type in what is the connection between smoking and depression? or I type smoking depression, I will not get very good results.
While it is true that some databases are getting “smarter” and know how to break apart a sentence (for example Britannica Online or the web search engine Ask Jeeves), in general it is not a good idea to search like this.
Every database has different rules about how they interpret what you type. Some will ignore certain words altogether – these are called stop words, and are usually very common words that would take up too many system resources if the database were to search for them. Others will allow you to search for two words together as a phrase, but will break apart any phrase you type that is longer than three words. It is a good idea to understand how the database you are using works before searching. This information is typically found in the Help screens.
Boolean, or “logical”, operators let you combine multiple search terms into a single search statement by using the connector words AND, OR, and NOT.
AND
- Use AND to find records that contain specific words or phrases.
- AND narrows your search to retrieve fewer, more relevant records.
- Use AND to combine two or more unlike terms.
- Your search results will contain both terms.
Boolean concepts are often explained with Venn diagrams – circles which represent sets of data containing the specified terms.
For example:
Search = smoking
This set of search results contains information only on
smoking.
Search = depression
This set of search results contains information only on
depression.
Search = smoking AND depression
This set of search results will find information on both
smoking AND depression. This essentially says, “I want to find information on the
connection between smoking and depression.”
Using AND can be confusing, because people often think
that the word “and” means to add things together to make a larger whole, however
using AND actually narrows your search and gives you fewer results.
OR
- Use OR to find records that contain one or more specific words or phrases.
- OR broadens your search to retrieve more items. “Or is more.”
- Use OR to combine two similar terms.
- Your search results will contain either of the terms.
Search = smoking or nicotine
This set of search results will find information on either
smoking OR nicotine. This says, “I want to find information on smoking or nicotine
– either word is relevant to my topic.”
Using OR can be confusing, because people often think that
the word “or” means to choose only one thing among many to make a smaller whole,
however using OR actually broadens your search and gives you more results.
NOT/AND NOT
- Use NOT to find records that contain the first specified word or phrase but not the second. Not removes any information from the search results that contains the second word.
- NOT narrows your search to retrieve fewer, more relevant records.
- NOT is used to exclude a certain term that might otherwise appear in your search results.
- NOT tends to be used less frequently than AND or OR, but it can be very useful.
Search = smoking NOT marijuana
This set of search results will find information on smoking
but NOT marijuana. This essentially says, “I want to find information on smoking, but
not about smoking marijuana.”
You have to be careful using NOT. The article's title or
abstract may mention marijuana anyway, saying it is not the focus of the article, but that
smoking cigarettes is. The article may be relevant to your search, but it will not be
retrieved using the terms, "smoking NOT marijuana."
COMBINATIONS
- You can combine Boolean expressions to make very sophisticated search statements.
- Use parentheses to group search statements.
Search = (smoking OR nicotine) AND depression
This set of search results will find information on either
smoking OR nicotine combined with (AND) depression. This says, “I want to find
information on the connection between either smoking or nicotine (because they’re
pretty much the same thing) and depression.”
Most databases observe a hierarchy or precedence with respect to operators. For example, the OR operator may be processed first in some databases; the AND operator in other databases.
Consider this:
If the AND operator is processed first, the search statement dogs OR cats AND leashes will find records that contain both cats and leashes or articles just about dogs.
If the OR operator is processed first, this search statement will find articles about either dogs or cats combined with the concept of leashes.
Note the difference.
To get around this ambiguity, always group in parentheses the parts of the search statement you want processed together. This is often referred to as nesting.
So, if we want an article about the use of leashes on either dogs or cats, we would structure the search as (dogs OR cats) AND leashes. We could include a related concept for leashes and expand the search like this: (dogs OR cats) AND (leashes OR harnesses).
Proximity operators are similar to Boolean operators. Proximity operators allow you to specify the relationship between multiple search terms by using the operators W and N. They are most useful when searching the full text of an article.
W (WITHIN)
- Use W plus a number to find records that contain a specified word or phrase within that specified number of words in the order you entered them.
- W narrows your search to retrieve fewer, more relevant records.
For example, "careers W5 computer" will find records with the word computer within 5 words of the word careers. Thus, this search would find “careers in computer science”, “ careers in computer software”, “careers requiring computer skills”, etc.
N NEAR
- Use N plus a number to find records that contain a specified word or phrase within that specified number in any order that you entered them.
- For example, "military N5 intelligence will find records with the word military near (either side of) intelligence. This search would find "intelligence operations in the military,"military information from intelligence sources."
TRUNCATION CHARACTER
Most databases specify a character that can be used to search for words containing a common word root, with any number or combination of characters following the root.
The truncation character in Academic Search Premier is "*"
For example, operat* will find records containing operations, operational, and operator.
Truncation can be useful when you don’t know whether or not your search term will appear in the singular or the plural, for example wildcat* will find wildcat or wildcats.
WILDCARD CHARACTER
Most databases specify a character that can be used to search for a specific number of characters, in any combination. You use one wildcard character for each character that you’re unsure of.
The wildcard character in Academic Search Premier is "?"
For example, wom?n will find records containing woman, women, and womyn.
All of these advanced searching techniques can be helpful in
streamlining your search for information. Once familiar with them, each researcher
develops his or her own set of preferred search strategies.
Provided that you are using the techniques correctly, there is no right or wrong way to search. The most important aspect of your search is the quality of the results you retrieve — are you finding information on your topic? Are your results precise, that is, is the information you found focused, or is there a lot of “junk” mixed in with the relevant information? The ideal search is one that is narrow enough to exclude extraneous or unrelated articles yet broad enough to include the most relevant articles.
You should continually keep an eye on the quality of your search results and adjust your search strategy accordingly. This means you should:
- Brainstorm an adequate number of keywords related to your topic.
- Check to see what controlled vocabulary terms (subject headings /descriptors) the database uses for your topic.
- Try new arrangements of your search terms using Boolean and proximity operators and nest your search terms in parentheses as appropriate.
- Broaden or narrow your search by specifying the fields to be searched.
- Place phrases in quotation marks to make sure they are searched as a phrase.
- Broaden your search by using truncation or wildcards if necessary.
NOTE: the Help feature available in most databases can
explain to you how to make the most of the database, thus ensuring that your searches are
efficient and your results are on target.
Copyright © 1994-2002. University of Rhode Island. Disclaimer.
Page last updated on Thursday, August 08, 2002