Skip to main content
OCLC Support

About searching WorldCat interactively

Discover how to search WorldCat in Connexion client.

Search for words, phrases, or whole phrases

Searching and browsing using word, phrase, and whole phrase indexes are available for most keyword, numeric, and browse indexes.


  • Title (ti:) [word index]
  • Title Phrase (ti=)
  • Title Whole Phrase (tiw=)

The following table shows how Connexion searches an index for words, phrases, and whole phrases and gives examples and guidelines:

Search for words, phrases, or whole phrases - Table
Index Matches search against... Examples/Guidelines for command line search area*
(Includes both keyword and numeric terms)

Use a colon (:)
Individual words (or numeric term) anywhere in an indexed field Publisher: pd:dakota

Dewey Class Number: dd:616

LC Class Number: lc:Z5074.G8
Enter multiple words to a single index.
Enclose the words in marks to match the exact sequence of words
Use with (w) or near (n) with number (1 to 25) to match the sequence of words with a specified number of words between
  • ti:asian crisis - Finds any instance of the presence of both terms in the title index.
  • ti:"asian crisis" - Finds only instances of the exact term in the title index.
  • ti:asian w2 crisis - Finds nstances of the two terms when the first term precedes the second term within two words.

Use an equal sign (=)
Multiple words in sequence within a single subfield, beginning with the first word in the subfield and including all data in the subfield Personal Name Phrase: pn=james, henry

Publisher Phrase: pb=namgyal institute of tibetology
  • Enter multiple words in order from the beginning of the first subfield of an indexed field.
  • Enter all data from the subfield or truncate using the asterisk (*). Alternative: Use a browse search in the Command Line. It is the only type of search that automatically truncates phrases.
  • Omit initial articles in titles.
Whole Phrase

Use an equal sign (=)

Whole phrase index labels end with the letter w.
Multiple words in sequence within a single field, beginning with the first word in the field, across multiple subfields, and including all data in the field. Personal Name Whole Phrase: pnw=james, henry 1843-1916

Subject Whole Phrase: suw=library science -- computer network resources
  • Enter multiple words in order from the beginning and include data from all subfields.
  • Truncate data using an asterisk (*). Alternative: Use a browse search in the Command Line. It is the only type of search that automatically truncates phrases.
  • Omit initial articles in titles.
*In the Keyword/Numeric Search area, omit index label and punctuation and select index from list.

 Caution: When you enter a phrase in the Command Line to browse, you must use an equal sign, not a colon. If you use a colon with a phrase, the system scans only for the first word of the phrase.


  • sca ti:gone with the wind browses only for the word gone
  • sca tiw=gone with the wind browses for the entire phrase

Use wildcards, truncation, and plurals

When searching keyword or numeric indexes, use the following symbols to broaden a search:

Symbol(s) Function Description Examples
# or ? Wildcard Use the # symbol or ? as wildcards (masking symbols) to match variant spellings:
  • Replace one unknown character with a # symbol.
  • Replace multiple unknown characters with the ? or ?n symbol (n = number of characters; use single digits 1 through 9).
  • adverti#e retrieves advertise and advertize
  • wom#n retrieves woman and women
  • computer? retrieves computer, computerization, or computerized
  • col?r or colo?r retrieves color or colour
  • int?net retrieves internet or intranet
* Truncate Use the * symbol to truncate a word or number to retrieve variant endings.
  • invest* retrieves investor, investing, investment fraud, etc.
+ Plural Use the + symbol at the end of a singular word to match plural forms ending in s or es. 
  • business+ retrieves business or businesses
  • lad+ retrieves lad or lads


  • Wildcard, truncation, and plural ending symbols must be preceded by at least three characters.
  • These symbols cannot be sued with non-Latin script search terms

Use Non-Latin script search terms

Connexion client supports using these non-Latin scripts for constructing WorldCat search terms: Arabic, Armenian, Bengali, Chinese, Cyrillic, Devanagari, Ethiopic, Greek, Hebrew, Japanese, Korean, Syriac, Tamil, and Thai.

The following are search guidelines for using non-Latin script terms:

Non-Latin script term search guidelines
  • Search for records containing non-Latin script data using either script data or romanized (Latin-script equivalent) data.
  • Both the Search WorldCat window (Cataloging > Search > WorldCat or <F2>) and the Enter Bibliographic Batch Search Key window (Batch > Enter Bibliographic Search Keys or <Alt><B><B>) support entering searches in non- Latin scripts.
  • Copy and paste non-Latin scripts into the Search windows or the WorldCat quick search box from any Unicode-compliant file.
  • Use the same WorldCat indexes (entering labels in Latin script) to search for both Latin and non-Latin script data. There are no separate indexes or index labels for retrieving non-Latin script records.
  • Add the same qualifiers (using Latin script) to both Latin and non-Latin script searches.
  • Generally, do not use truncation, wildcards, or plural ending symbols with non- Latin script searches. An exception to this guideline is to use truncation for the following scripts:
    • Bengali
    • CJK (Chinese, Japanese, and Korean)
    • Devanagar
    • Tamil
    • Thai
  • Do not use non-Latin scripts in derived searches.
  • See more details on display and sort order of search results that include records with non-Latin scripts in International cataloging.
  • Scripts available are:
    • Arabic
    • Armenian
    • Bengali
    • CJK (Chinese, Japanese, and Korean)
    • Cyrillic
    • Devanagari
    • Ethiopic
    • Greek
    • Hebrew
    • Syriac
    • Tamil
    • Thai
  • To view records containing a particular non-Latin script, use the index label vp: and the script designation:
     Note: View non-Latin scripts only in the Connexion client. You cannot view multiscripts in the Connexion browser, unless you save the records in the online save file using the Connexion client and then open the save file records in the browser. The browser display of non-Latin scripts in the online save file is read-only.
    • vp:ara
    • vp:arm
    • vp:ben
    • vp:cjk
    • vp:cyr
    • vp:dev
    • vp:eth
    • vp:gre
    • vp:hbr
    • vp:syr
    • vp:tam
    • vp:tha

Tips for catalogers working with WorldCat Arabic and/or Persian records

Navigate to Tools > Options > International to automatically transliterate romanized data into Arabic script data for Arabic and Persian WorldCat records when you retrieve them from WorldCat interactively (does not work for batch searching). The client automatically transliterates records that have the language code ara or per but no field 066. Also, select the fields you want auto-transliterated. In the International window (Tools > Options), click Choose Fields.
Transliterate manually field by field by clicking Edit > Transliterate > Arabic [or Persian].

Tips for catalogers working with CJK scripts

Click Edit > MARC-8 Characters > Convert to MARC-8 CJK to automatically convert any invalid CJK characters in a record.

Use the built-in CJK E-Dictionary (Tools menu) to help select characters.