A specialized database—often called a research or library database—allows targeted searching on one or more specific subject areas (i.e., engineering, medicine, business, etc.), for a specific format (i.e., books, articles, conference proceedings, video, images), or for a specific date range during which the information was published. Specialized databases can provide you an index of scholarly information available as well as full-text access to that material.

There are several types of specialized databases, including:

  • Bibliographic – details about published works
  • Full-text – details plus the complete text of the items
  • Multimedia – various types of media, such as images, audio clips, or video excerpts
  • Directory – brief, factual information
  • Numeric – data sources
  • Product – model numbers, descriptions, etc.
  • Mixed – a combination of other types, such as multimedia and full-text

When to Use Specialized Databases

Search specialized databases to uncover scholarly information that is not available through a regular web search. Specialized databases are especially helpful if you require a specific format or up-to-date, scholarly information on a specific topic.

Many databases are available both in a free version and in a subscription version. Your affiliation with NJIT allows you access to subscription-based databases at no cost to you. For example, using the Scopus database (a bibliographic database) via NJIT Libraries enables a Find It link to help you access an item that is available from other subscription databases.

Tip: Free vs. Subscription?

In some cases, the data available in free and subscription versions are the same, but the subscription version provides some sort of added value or enhancement for searching or viewing items.

Database Scope

Information about the specific subject range, format, or date range a particular specialized database covers is called its scope. A specialized database may be narrow or broad in scope, depending on whether it, for instance, contains materials on one or many subject areas. Using the NJIT Library website, click on the Databases tab at the top of the main search box, then click the link for Databases by Subject to find the database with the particular scope you need.

For more information on how to access the NJIT Library databases and how to determine which database to search:

See the following Research Guide, Using Databases


How to Use Them

The use of each database varies somewhat.

See NJIT’s research database list in an alphabetical list or by subject.

Example: Academic Search Premier

Academic Search Premier obtained through EBSCOHost (NJIT only) is a general article database available through most academic and large public libraries that is often recommended for undergraduate research projects.

Movie: Academic Search Premier Database in less than 3 Minutes

View video

Keyword Searching

Although keyword search principles apply (as described in Precision Searching), you may want to use fewer search terms since the optimal number of terms is related to database size. Google works best with several terms since they index billions of web pages and additional terms help narrow the results. Each scholarly database indexes a fraction of that number, so you are less likely to be overwhelmed by results even with one or two keywords than you would be with a search engine.

Phrase searching (putting multiple words in quotes so Google will know to search them as a phrase) is also less helpful in specialized databases because they are smaller and more focused. Databases are better searched by beginning with only a few general search terms, reviewing your results and, if necessary, limiting them in some logical way. (See Limiting Your Search below.)

Activity: Compare Them!

Instructions:
Compare a search for items containing both phrases “United States” and “female serial killers” in the article database Academic Search Complete (NJIT only) and in the web search engine Google Scholar. (Make sure you include the quotation marks so they will be searched as phrases.) Notice how searching too narrowly (searching for phrases) affects results in the specialized database. How could you revise the specialized database search to get more results?

Limiting Your Search

Many databases allow you to choose which areas (also called fields) of items to search for your search term(s), based on what you think will turn up documents that are most helpful.

For instance, you may think the items most likely to help you are those whose titles contain your search term(s). In that case, your search would not show you any records for items whose titles do not have your term(s). Or maybe you would want to see only records for items whose abstracts contain the term(s).

When this feature is available, directing your search to particular parts of items, you are said to be able to “limit” your search. You are limiting your search to only item parts that you think will have the biggest pay-off at distinguishing helpful items from unhelpful items.

Searching fields such as title, abstracts, and subject classification often gives helpful items.

Tip: Full-Text Searches

Some databases allow for full-text searching, but this option includes results where a search term appears only once in dozens or more pages. Searching fields such as title, abstracts, and subject classification will often give more relevant items than full-text searching.

Subject Heading Searching

One precision searching technique may be helpful in databases that allow it, and that’s subject heading searching. Subject heading searching can be much more precise than keyword searching because you are sure to retrieve only your intended concept.

Subject searching is helpful in situations such as:

  • There are multiple terms for the same topic you’re interested in (example: cats and felines).
  • There are multiple meanings for the same word (example: cookie the food and cookie the computer term).
  • There are terms used by professionals and terms used by the general public, including slang or shortened terms (example: flu and influenza).

Here’s how it works:

Database creators work with a defined list of subject headings, which is sometimes called a controlled vocabulary. That means the creators have defined which subject terms are acceptable and assigned only those words to the items it contains. The resulting list of terms is often referred to as a thesaurus. When done thoroughly, a thesaurus will not only list acceptable subject headings but will also indicate related terms, broader terms, and narrower terms for a concept.

Tip: Finding Useful Subject Headings

Try this strategy to find useful subject headings. Remember it by thinking of the letters KISS:

  • Keyword-search your topic.
  • Identify a relevant item from the results.
  • Select subject terms relevant to your topic from that item’s subject heading.
  • Search using these subject terms. (Some resources will allow you to simply click on those subject terms to perform a search. Others may require you to copy/paste a subject term[s] into a search box and choose a subject field.)

Records and Fields

The information researchers usually see first after searching a database is the “records” for items contained in the database that also match what was asked for by the search.

Each record describes an item that can be retrieved and gives you enough information so that, hopefully, you can decide whether it should meet your information need. The descriptions are in categories that provide different types of information about the item. These categories are called “fields.” Some fields may be empty of information for some items, and the fields that are available depend on the type of database.

Example: Database Fields

A bibliographic database describes items such as articles, books, conference papers, etc. Common fields found in bibliographic database records are:

  • Author.
  • Title (of book, article, etc.).
  • Source title (journal title, conference name, etc.).
  • Date.
  • Volume/issue.
  • Pages.
  • Abstract.
  • Descriptive or subject terms.

In contrast, a product database record might contain the following fields:

  • Product Name.
  • Product Code number.
  • Color.
  • Price.
  • Amount in Stock.

License

Icon for the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License

Choosing & Using Sources: A Guide to Academic Research Copyright © 2015 by Teaching & Learning, Ohio State University Libraries is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License, except where otherwise noted.

Share This Book