/
Search Term Codebook

Search Term Codebook

Purpose: Following is an explanation of fields and terms used for coding the search queries obtained from Encore. The purpose is to better analyze the ways that users are engaging with the catalog and gain a sense of how users are searching for information. General procedures for assigning codes to each query are described first.  Definitions of each field and term are provided in the “Fields” section. 


Tools Needed 

  • Airtable 

  • Google 



Outline of Procedures 

 Open the Time Sorted view in the URLs tab of the Discoverability Research project in Airtable 

  1. Copy and paste the search terms from the “Scraped Search Terms” field into Google to gain a broader sense of what the user might be trying to research 

  2. Assign a Search ID for all searches that appear to be done by a single user in the Search ID field 

  3. Assigned a synthesized search topic in the “Synthesized Search Term” field  

    1. Note: ALL lines sharing a search ID should have the same synthesized search terms -  more details on how to do this below 

  4. Indicate which search type describes the query in the “Search Type” field 

    1. If a known item is selected, indicate which kind of known item in the “Known Item Type” field 

  5. Assign the type of keyword used in the Keyword Sequence field.  If more than one type is used, assign them in the order they occur. 

  6. IF applicable, assign a category for additional details for the search terms used in the “Keyword Details” field.  See below for details.   


Fields 

The following information will outline how each field should be coded.  Please note that the Search ID and the Synthesized Search Topic are used to group together searches that appear to be done by single user.  All rows with the same search ID should include the same Synthesized search topic. 

 All other fields will be coded based on the actual search terms used and not on the synthesized search terms.   


Search ID 

 Definition: This field is used to group together searches that appear to be done by a single user.  A new ID is issued for each new “search group”.   

 Structure of Code: IDs begin with the letter “S” followed by four numbers.  They run consecutively (S0001, S0002, S0003) 

 Instructions and Examples: A search group is composed of any search strings that fits two criteria:  

  1. Appears to be related to other searches (Note: you will likely need to conduct a general google search to gain a sense of potential related terms) 

    1. Example 1: the search for “Teaching Cues for Sport Skills” is very closely related to “Teach Cues for Sports Skills fronkse” because the user clearly added an author to the end of a search for a title.   

    2. Example 2: the search for “Nin, Anaïs, 1903-1977.” is very closely related to “delta of venus” and “little birds” because the first is an author search and the second and third are titles of her book. 

  2. Occurs within 30 minutes of a related search 

 

Purpose: To be able to group search query sets together and sort by them 


Synthesized Search Topic 

Definition: A brief set of terms that summarize the main topic for which the user was searching.   

Structure of Code: 1-3 terms (where possible).  Copy the user’s search terms wherever possible. 

Instructions and Examples: All items within a single Search ID group should have the same term.   

For example a search group with the following search terms: “WEBA,” “WEBA political group,” “women on web,” “aid access,” “women on web abortion,” “aid access abortion,” “medical abortion procedures” was assigned the synthesized search topic “abortion.”   

 

Purpose: To aid in grouping search topics together and getting a broader scope of potential areas of interest for patrons 


Search Types 

Definition: The category that explains the type of search query employed by the user. 

Structure of Code: Pre-selected list 

  • Known Item: Search term(s) that uses any citation information to find a specific item (usually an article or book) 

    • Example: “valuation and risk management of energy market” used to find the book Valuation and Risk Management in Energy Markets 

    • Example: “ngai impossible” used to find the book “Impossible Subjects by Mae Ngai 

  • Resource Search: Search term(s) intended to find a database or library resource that is not a specific item 

    • Example: “CQ Researcher” or “Credo Reference” 

  • General Topic: Search term(s) that are used to find information on a topic, but are not specific to any known item, person, organization/entity, etc. 

    • Example: “Abortion” or “Belonging in groups” 

  • Person: Search term(s) that indicate the name of an individual, where it is unknown or unclear if the search is intended to find information ABOUT the individual or information CREATED by the individual 

    • Example: “John Steiner” 

  • Organization/Entity: Search term(s) that indicate the name of an organization, where it is unknown or unclear if the search is intended to find information ABOUT the organization or information CREATED by the organization 

    • Example: “American Pregnancy Association” 

  • Idea Generation: Search term(s) that indicate a need to find potential research topics 

    • Example: “Early childhood research topics” 

  • Specific/Named Topic: Search term(s) that indicate a topic, where it is unknown or unclear if the search is intended to find information ABOUT the named topic or the named topic itself 

    • Example: “Shanghai International Settlement” 

  • Format: Search term(s) that include a specific format  

    • Example: “sam cooke sheet music” 

  • Error: Search term(s) that appear to be typed in error. Do not include spelling mistakes. 

    • Example: “https:” 

  • URL: Search terms that are web addresses  

 Instructions and Examples: Select the most appropriate search type term from the list of options.  Select all that apply.  Where appropriate, put multiple terms in order of their appearance.   

For example: “sam cooke sheet music” has the Search Types: Person, Format 

 

Purpose: Categorizes the searches to allow for general statistical analysis on they types of searches used most frequently, or most frequently paired together. 


Known Item Type

 Definition: The format of the material for which the user was searching, applicable to known items only.  

 Structure of Code: Select from the drop-down menu of format types. 

 Instructions and Examples: ONLY fill out this field, if the Search Type is a “Known Item”.  Leave it blank if it is not a known item. More than one item may be selected from the list if the known items is in multiple formats and it is unclear which format the user preferred. 

 

Purpose: To determine what kinds of materials are searched anticipated by users.


Keyword Sequence

Definition: The structure of the search terms used by the patron. 

Structure of Code: Pre-selected list (note: this list will expand) 

  • Creator 

  • Creator/Subject term 

  • Database 

  • DOI 

  • Edition 

  • Format 

  • General topic term(s) 

  • Journal 

  • No 

  • Pages 

  • Person 

  • Question 

  • Specific/Named topic terms(s) 

  • Title 

  • Translation 

  • Volume 

  • Year 

 

Instructions and Examples: Add all of the keyword construction types that apply to the original “Scraped Search Term”.  Put them in the order they appear. 

For example:  a search of “Cole, Holly E., and Denise R.  Beike. “Tall Tales Make Fast Friends: Exaggerating When Retelling Previous  Experiences Fosters Relational Closeness.” Journal of Social and Personal  Relationships, vol. 36, no. 8, Aug. 2019, pp. 2287–2306. EBSCOhost, doi:10.1177/0265407518787344.”  would have the following sequence: Creator, Title, Journal, Volume, No, Year, Pages, Database, DOI 

 

Purpose: This field will help to analyze the general construction of keywords, including what elements users employ, the typical ordering of keyword element, or how often certain pairing of keywords occur. 


Keyword Details

Definition: Additional information about the search, divided into categories that will be useful for analysis.  This list can be expanded, as needed. 

Structure of Code: Pre-selected list 

  • Citation Format: Search terms that appear to be citations, in part or in full. 

    • Example: Clarke, Colin. “Politics,  Violence and Drugs in Kingston, Jamaica.” 

    • Example:  Cole, Holly E., and Denise R.  Beike. “Tall Tales Make Fast Friends: Exaggerating When Retelling Previous  Experiences Fosters Relational Closeness.” Journal of Social and Personal  Relationships, vol. 36, no. 8, Aug. 2019, pp. 2287–2306. EBSCOhost, doi:10.1177/0265407518787344 

  • LoC Construction: When the search term is done in the same order as Library of Congress name or subject term.  It may or may not include the birth/death date. 

    • Example: Nin, Anaïs, 1903-1977. 

    • Example: L'Engle, Madeleine 

  • USU Faculty: When a search term appears to be as search for a USU faculty member 

    • Example: “shouqing si” 

  • Question: When the search term is presented as a question.  It may or may not include punctuation 

    • Example: “should prostitution be legal” 

    • Example: "Is the US Justice System Corrupt?" 

  • Idea Generation: When a search term(s) appears to be looking for ideas for research topics 

    • Example: Early childhood education research topics 

 

Instructions and Examples:  Only fill out this field, if it applies to the search terms.  

 

Purpose:  This field will be used as a catch-all to code for patterns that are seen that are not included in the other coded fields. 

Related content