MARS Authority Control

Any time we process a file for your institution, we generate a Statistical Summary report.  Within this report, we list match-rates, types of records processed, as well as the number of fields changed during processing.  While this is all well and good, as soon as that report is sent out to you, those statistics have been forgotten.

We thought it would be a good idea to keep you informed of some of those statistics on a regular basis, all completely tailored to when you send us files for processing.  And we wanted to do this in a graphical way so that you’re not just looking at lists of numbers and confused as to their relevancy to your records.

As we started planning on what to do with the statistics, we realized early on that not everyone goes through our website to process their files.  Some of our clients use external FTP servers, while some use the website.  We never want our customers to feel left out on any functionality we have implemented.

Whenever a file is submitted to Backstage for processing, we keep track of certain statistics within our tables.  We have to keep track because this is how we generate that Statistical Summary report that we deliver for base files, ongoing current cataloging, and samples.  It doesn’t matter how the file reaches us (website, ftp, email, etc)—once we process it, we keep track of the data.

This statistical tracking helps us determine the busiest times of the year and how we should structure the time on our side to meet the expectations on your side.  We also realized that this could be a useful tool for you and your staff to track the trends in your own processing.  Throughout the year, how often do you send in files for processing?  What kinds of match-rates or types of records do you usually see?

We have created two sections for you:

1. Record Statistics
2. Match Statistics

For Record Statistics, we’ll show you how many books, serials, and other types of records were processed:

Of those records, how many were changed vs unchanged?

We’ll also plot how many records you sent throughout the year:

We think it makes sense to only include months where you send us bib files to process.  Each part of every graph has parts that can be clicked to determine more information, or isolate that part of the process for you.

With Match Statistics, we go a step further than our Statistical Summary report and tell you about the kinds of matches you’ve received on your records:

Finally, we wanted to also show you how many of your headings (1xx, 6xx, 7xx) found full or partial matches, as well as the number of unmatched headings:

Over the next few weeks, we plan on putting this information in place for all of our clients.  Whenever you log into our website to view your profile or current jobs, you can also view your stats.  Again, it doesn’t matter whether you send us your bibs through the website, ftp server, or email—once we process them, those stats will be added to your stats page.

As a bonus to those clients that do send us their files through the website, each one of those website bibliographic jobs will have its own set of graphic statistics as well.

We hope this information gives you the edge you may need for your staff.  As always, feel free to let us know if you would like to see other statistics.

New Optional Report – R08 Generic Names

Backstage has a new optional report that can be generated during Current Cataloging or Basefile processing.  The new report lists all generic name headings.  Below is a brief description of the report as well as a few examples of how headings on this report appear.

The fields in this report contain a primary name heading which is considered generic (i.e. names without qualifiers such as a date or fuller form of the name). Choosing to flip on generic name headings will give you a greater match rate, but it can cause mismatched headings. The MARS 2.0 software allows for the flip of all name authority headings including the generic heading if there is a match to a See reference (400) in the LC authority record.

Sometimes a national authority record just lacks some of the key information that your library needs. Maybe it’s an important cross-reference, a note, or some sort of field identifier. If you just receive the latest Library of Congress authority, for example, you could lose all that hard work that you and your staff have gone through to find the information and add it to your authorities.

Thanks to our project manager for writing that up. Preserving fields is one of the services many of our libraries find most useful, and we have even more options than listed above when it comes to preserving your data for cases where adding a $5 would be unreasonable or impossible. so feel free to talk to on of our project managers about how we can help you with your situation.

With all the excitement surrounding RDA at ALA Annual (and we’ll have an update about that up soon), we realized we didn’t have a quick recap of events from my point of view in DC.  I would like to fill you in on details from an operational point of view as concerns Backstage, letting Karen Anderson let you know about the RDA updates later.

On June 24, a few of us flew into Bethlehem, PA to visit our new offices out there.  As some of you may know, we acquired the preservation services (Digitization, Microfilm) from OCLC back in August 2009.  So it was an opportunity to get out and visit with our team, meeting Joan DaShiell and Joe Konrath, both of whom are widely respected in that business.

That next Friday we drove up to DC to take part in the sweltering outside heat and air-conditioned halls of the conference center.  The traffic going into DC made the trip an all-day affair, though the upside was it gave us a chance to catch up on our emails and plan out our schedules.

On Saturday, we had a great turnout for the Authority Control users breakfast as Clyde’s.  I mentioned a couple announcements (i.e., online stats to track your processing throughout the year, and our ability to process & return records in MARCXML now).  But the real star of the show was with Karen’s updates regarding RDA.  As Backstage is part of the ongoing testing effort with RDA, it was great to have Karen with us to field your questions.  She has written up a detailed response and we will post that very soon.

Much of the remainder of my time was occupied with chances to meet up with existing and potential clients.  It’s a great feeling to work with people all throughout the years and finally meet each one in person.

We think we have a fair number of lurkers on this blog and would encourage you to leave comments whenever you like.  We are always on the lookout to make the service easier and more efficient for you and your workflow, so please feel free to let us know how we can make things better.  Tell us what you want and we will see what we can do to make it happen.

Why is Authority Control Important?

A basic definition of Authority Control is the process used to maintain consistency in access points and the process of showing the relationships among names, subjects, and works.

Some basic goals of authority control go back to Charles Cutter’s goals for a library catalog:

• To enable a person to find a book
• To show what a library has in its collection
• To assist patrons in their choice of a book

Using Cutter’s goals for a catalog, we can go further by saying that the purposes of authority control are to help patrons:

• find an item from known information
• identify an item with limited information
• collocate items using a controlled vocabulary
• evaluate and select an item
• locate synonyms to aid subject searching

To achieve these goals, authority records need to bring together related terms for the specific name, subject, or work.   Many times, the name of a person, place, or thing is constantly changing.  For example, the artist currently known as “Diddy” has had several name changes over the years.  Without a current authority record, a patron may not be able to find all of the items by this artist if they are searching on his former name of Sean Combs.  With the following authority record, the patron would be able to search for any of the artist’s former names and find all relevant items.

010 $a no 97057848

100 $a Diddy, $d 1969-

400 $a Puff Daddy, $d 1969-

400 $a P. Diddy, $d 1969-

400 $a Combs, Sean, $d 1969-

400 $a Combs, Puffy, $d 1969-

400 $a Puffy, $d 1969-

400 $a Diddy, P., $d 1969-

400 $a Combs, $d 1969-

400 $a Combs, Diddy, $d 1969-

Authority control is also important when a well-known name has been translated into several different languages.  For instance, the authority record for William Shakespeare has ninety-seven alternate forms of his name translated into multiple languages.

In order to increase access for the library’s patrons, it is important to use authority control to standardize the forms of search terms and increase the likelihood that the search will return the items that the patron desires.

Reference:

Taylor, A. G., & Joudrey, D. N. (2009). The Organization of Information (3rd ed.). Westport, Conn.: Libraries Unlimited.

We’re excited to announce that the 300 field validation that we’ve been working on is now in place and working.  We add this to the hundreds of other corrections and validations that we make to bibliographic records as we process them. Judy, one of our Project Managers, explains the new process:

In an effort to continue improving our bibliographic validation services, we’ve been writing some cleanup steps for the 300 field.  Nothing drastic of course — just some “straightening out” changes.  The cleanup takes place only on records marked with ‘a’ in leader byte 6 … non-print bibs such as media, electronic resources, microforms, etc. are left as-is.  Here’s a list of what the programming will do:

• * 300 $a = ‘unp.’ [or] ‘n.p.’ is changed to ’1 v. (unpaged)’
• 300 $a = a comma is added to #### page numbers if missing (ex.: 5332 p. becomes 5,332 p.)
• 300 $a = ‘l.’ is changed to ‘leaves’
• 300 $a = ‘approx.’ is changed to ‘ca.’
• 300 $a = preliminary leaves and it’s abbreviation (p.l.) are removed (ex.: 4 p.l., xv, 312 p. = xv, 312 p.)
• 300 $a = ‘unnumbered’ is removed and brackets added (ex.: 48 unnumbered leaves = [48] leaves)
• If the p. is up against the :$b or the illustrative matter is up against the ;$c we add a space to separate
• 300 $a = Missing periods are added to ‘p’ (ex.: 135p. = 135 p. )
• 300 $b = The illustrative matter is rearranged to be in correct order
• 300 $b = Obsolete illustrative matter is replaced with ‘ill.’ (ex.: ‘tables’ or “fronts.’ = ‘ill.’)
• 300 $b = If ‘ill.’ is followed by the parenthetical illustrative matter, ‘ill.’ is dropped and the parenthetical matter remains (example: $bill. (ports.) = $bports.)
• 300 $b = illus. is changed to ill.
• Adds missing semicolon before $c
• Adds ;$ccm. if missing from 300 field
• [various pagings] = changes brackets to parentheses
• If there is no 300 and the record is a book, the default of $ap. ;$ccm. is added
• If there is numbered plate info in the $b that is moved to the $a – if plate designation is not numbered, it’s deleted

As with any of validation rules, these are optional, and if necessary, we can work with you to develop exactly what you’d like. These rules are now being included by default in our processing. We hope you find them useful!

Last week we talked a little bit about the history of how series have been handled, but what is series authority control? Of all the types of Authority Control (Names, Subjects, Series) series is likely the most complex aspect and can be quite overwhelming for both newcomers and veterans of Authority Control. Our Authority Librarian, Karen, has once again come to our aid to help explain what a series is, and how Authority Control works with series.

A series is defined by AACR2 as a group of separate items related to one another by the fact that each bears, in addition to its own title proper, a collective title applying to the group as a whole.  A series under authority control uses a controlled or unique chosen access point for the collective title of a series in the analytic bibliographic records.

Series authority work is the process of controlling those collective titles through the use of authority records.  Series authority work is done for the same reasons as name authority work; that is first, to create a single form of the series title which will collocate records so that the user can find all the records related to that resource; and second to create a unique form of the series title which will distinguish records for one resource from those for a different resource.  Uniqueness is especially important since many series have the same or similar titles.

There are several decisions that need to be made to control a series.  These decisions should be based on a standard such as AACR2.

The first decision is to decide which form of the series title will be used as the authorized form.  A single, unique form of the title needs to be chosen often from among several variant forms.  Then that decision needs to be recorded.  Three possibilities for this include:

• Use the bibliographic record for the resource used for the series data to record the authorized form.
• Create an alphabetic listing of series in the library.
• Use a Series Authority Record  (SAR)

The SAR provides the best method for access, and can also contain a lot of other valuable information about the series.  The authority record can be on a card, or most commonly now, a machine-readable record.  More on SAR’s below.

The second decision to make is how the series will be treated.  How will you handle the parts of the series that come in to your institution?  A couple of points to consider are:

• Analyze or not
• Class together or separately

The third decision is whether to trace the title or not.  This means deciding whether to use the access point in your database or catalog.

It is important that these decisions be recorded because not all series will be handled in the same way.  Actions taken for subsequent acquisitions need to be consistent.  The most common way to record series authority records is a MARC 21 record format.  Since MARC21 is a machine-readable format, the records can be updated at any stage in the workflow.  They are based on standards and can be easily shared with the greater library community.

A Series Authority Record (SAR) is the means to generate consistency in any database (or catalog) and allow the retrieval of all the series in a logical display.  The SAR should serve three functions:

1. Provide a unique and authoritative heading.  The record reflects the chosen form that will be used as an access point for the series for all bibliographic records containing references to the series.  Uniqueness promotes consistency in the database and allows the retrieval of all analytics in a predictable display.  Several fields in a MARC format SAR provide the required information.  The 1xx field is used to record the chosen form of the series title.  This is the authorized heading.  Related to the function of the 1xx to record the chosen form, the 4xx fields are used to record variant forms of the heading (series title) that were not chosen as the authorized form.  The 642 field provides the style of numbering to be used with the series if it is numbered and also provides consistency.
2. Provide identification – the record should provide enough information to identify the resource used in doing the authority work for a particular series, and distinguish this series from all others.  The 1xx and 4xx fields are also part of this function.  Other fields providing identification are:

643 –   contains the place and publisher of the series

670 –   contains the time period of publication

642 –   numbering example taken from the source listed in the first 670 field.

667 –   this is a note field that is used to record other information that is valuable,  but does not fit into other fields.

1. Show the treatment – the SAR should be a record of the decisions regarding analysis, tracing, and classification.  It can have both local and other decisions.  Fields used for this function are:

050-090 –  These fields record the call number chosen for the series if the decision is to class holdings together.

644 –   this field records the analysis decision

645 –   this field records the tracing decision

646 –   this field records the classification decision (see fields 050-090)

One other type of SAR can actually be for something that is not a series.  Many resources have phrases or other character strings that will not be considered as a series.  Although a SAR is not required to record this conclusion, providing one with the documentation for the decision to not consider it a series will create consistency for questionable cases.  It will also save time in the future by avoiding having to make the same decision over and over again as other materials come in.

In a MARC21 bibliographic record the 490 field is used to record the transcription of the series from the item.  In this field, the series appears exactly like it does on the item.  A 500 note field may also be used for the series.  The 8xx fields (800, 810, 811, and 830) are used to provide a tracing for the series when needed.  These fields will contain the authorized form of the series title.  This demonstrates the need for the authority work to provide consistency.  The series on items do not always appear the same way so there could be different headings in the 490 fields for the same series.  But by having one authorized form, the 8xx fields for that series will always have the exact same heading.  Without this consistency, it could be very difficult to find all the parts of the same series in the catalog or database.

Series authority records can provide effective management for the bibliographic headings and can also provide physical control of series holdings in the library. By creating series authority records and maintaining a database of the records (an authority file) consistency will be maintained.  Time and effort will be saved in the future.  As new materials come in, the work on the series and the decisions regarding it will not have to be done over and over again.

For a detailed explanation of Series authority work see the excellent power point presentation by Judith A. Kuhagen of the Cataloging Policy & Support Office at the Library of Congress.  http://library.ust.hk/info/other/feb2005/series.html Although this presentation is from 2005 and some of the LC policies and MARC tags talked about have changed, the process of making the decisions needed for series authority work in gone into in great detail.

For more information on the changes to series policies and MARC tags see the article “A Brief History of Series Authority Work”

For more information on using the MARC21 standards, see the MARC21 Format for Authority Data http://www.loc.gov/marc/authority/ecadhome.html and the MARC 21 Format for Bibliographic Data http://www.loc.gov/marc/bibliographic/ecbdhome.html

Judy Archer shares with us her insight on genre authorities and GSAFD.

Ahhh genre terms … what a subject is, versus what a subject is about.  The necessity for this distinction seems obvious, doesn’t it?  If I want to find poetry written by Robert Frost, I wouldn’t necessarily want to weed through all the works written about his poems, would I?  The more logical option is to access only his actual poems.  Yet for a long time I had no choice, for genres were considered part of the general subject access.

As time went by, the library community saw the need for a distinction and recommendations were developed for a national standard.  This standard provided guidelines for developing genre access as well as subject access for fictional works, humorous works, drama, poetry and folklore.

To accommodate this effort, the “Guidelines on Subject Access to Individual Works of Fiction, Drama, etc.” 2^nd edition was published in 2000, and this came to be known as the GSAFD.  In chapter 1 of that Guideline was a list of 153 genre terms, and in 2001 the Association for Library Collections and Technical Services (ALCTS) developed MARC21 authority records for these 153 terms.  The source file is located at Northwestern University, but OCLC has a link to the terms and their full authority records, which can be accessed at the following link:

http://alcme.oclc.org/gsafd/OAIHandler?verb=ListIdentifiers&metadataPrefix=z39_19

If you would like to know more, The ALCTS discusses their creation of the authorities here:

http://www.ala.org/ala/mgrps/divs/alcts/resources/org/cat/marc21authority.cfm

Discussions continue about the formulation of genre terms, and the Library of Congress began their own ambitious project in 2007 to formulate authorities for genre terms.  They started with Moving Images, but they have also moved on to additional categories such as:  Cartography, Law, Religion, Literature and Music.   They have a timeline and commentary about the project which you can find here:

http://www.loc.gov/catdir/cpso/genretimeline.pdf

Many libraries will use terms from the GSAFD list, the Library of Congress authorities, or other thesauri.  If you plan to use the 153 GSAFD terms in your catalog, they are handled by way of a 655 field using a second indicator of ‘7’ and a subfield $2 with the term ‘gsafd’.  This clarifies the thesaurus you used.  Here are examples of how both GSAFD and LC terms are entered in MARC21:

From the GSAFD list:

            655_7  $aEpic poetry.$2gsafd

From the LC subject headings catalog:

            655_0  $aRoad films     [or]     655_7  $aRoad films.$2lcsh

Many people know what the Library of Congress Authority Record Database is and how it can help their library. But what do you know about the other large authority collections? Nate Cothran, Product Manager of MARS Automation Services and expert on the Art and Architecture Thesaurus (AAT) sheds a little light on it for us:

The Art & Architecture Thesaurus (AAT) is a collection of 131,000 authority records about the visual arts and architecture maintained by the Getty Institute. Though the AAT is updated every month, the licensed version—which Backstage uses—is sent out once annually.

Within the authority record itself, the AAT entry lists the main heading in the 155 tag:

001 __ 300015529

155 __ $abird's-eye views$B(<views by vantage point or orientation>, views,
<visual works by form: image form>, <visual works by form>, <visual works>,
Visual Works, Visual and Verbal Communication, Objects
Facet)$I1000015529$TDescriptor$GN/A$HCurrent$VUndetermined$L70052/American
English$AN/A$ON/A$8300015529.1

The 155 $a is the part of the heading that will be reflected in your updated bib heading:

 655 __ $abird's-eye views

I can see what you’re thinking on this: How is it that the authority’s 155 above contains so much more information than the bib’s 655 field? Sometime in the last year, the AAT authority records began including an incredible amount of reference points & additional information encoded into capital-letter & numbered subfields. Unless your system is able to incorporate this extra data, the pertinent information you likely prefer to retain resides in the main (lower-cased) subfields.

In order to bring the authority records received from AAT back into a more manageable structure, Backstage can remove all of the capital-letter subfields as well as the numbered [0-9] subfields. This makes the loading into your system error-free in case your system balks at the extra subfields it has to somehow categorize on the fly.

So this revised authority heading:

001 __ 300015529

155 __ $abird's-eye views

Now matches cleanly and updates easily within your ILS system.

At least now you’ll know what to look AAT in your Art & Architecture Thesaurus authorities.

http://www.getty.edu/research/conducting_research/vocabularies/aat/about.html