EID Tagging User Guide - ERA 2015 Citation Supplier and EID Tagging
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
EID Tagging User Guide ERA 2015 Citation Supplier and EID Tagging September 2014
1
Table of Contents
1. Introduction 2
2. Citation Data Supplier 2
3. Tagging of Publications for ERA 2
3.1 EID Tagging Period 3
3.2 EID Tagging Registration 3
3.3 EID Tagging Procedure 3
3.4 Research Outputs to be EID Tagged 4
3.5 File Formats for Publications to be EID Tagged 6
3.6 Turnaround Time 9
4. Data Provision to Universities 9
4.1 File Formats provided back to Institutions 10
4.2 EID Tagging Status Definitions 11
5. Contacting Elsevier regarding Publication EID Tagging 12
6. Contacting the ARC regarding Enquiries and Comments 12
Appendix A. Scopus Document Types and Characteristics 13
Appendix B. Examples of Correct and Incorrect Field Columns for EID Matching 152 1. Introduction This EID Tagging User Guide provides background information about the publication EID Tagging procedure during ERA 2015, as well as detailed instructions to Australian Higher Education Providers (HEPs) on how to submit publications for EID tagging. This EID tagging of publications is offered as a service to the HEPs by the Australian Research Council (ARC), as a preparatory step prior to submission of publications for ERA 2015 (see also “ERA 2015 Submission Guidelines” at http://www.arc.gov.au/era/era_2015/2015_keydocs.htm). 2. Citation Data Supplier Scopus, Elsevier’s abstract and citation database, is the sole bibliometric and citation data supplier for the 2015 Excellence in Research for Australia assessment. 3. Tagging of Publications for ERA To be eligible for citation analysis, all articles must have an EID (or Electronic ID), which is a unique electronic identifier for each article indexed in Scopus. The EID allows quick and accurate assignment of citation data and appropriate benchmarks by the ARC post-submission. In preparation for submission, institutions will need to obtain an EID for each eligible journal article from an ERA citation discipline that is published in a journal indexed by Scopus. To facilitate this, the ARC has made arrangements with Elsevier to undertake EID tagging for each of the submitting institutions. The cost of the EID tagging will be borne by the ARC. Frequently Asked Questions What is an indexed journal? A journal is considered ‘indexed’ when the bibliographic and citation information is included by the citation data supplier, in this case Scopus. How do I know whether a journal is indexed? To find out whether a journal from the ERA 2015 list is indexed in Scopus, and for which coverage years, please look up the journal in the Scopus journal title list, which can be downloaded as an Excel file here.
3
3.1 EID Tagging Period
Institutions will be able to submit publications for EID tagging directly to Elsevier from Midnight AEST
(Australian Eastern Standard Time), Wednesday 1 October, 2014, until 17:00 AEDT (Australian
Eastern Daylight Time), Friday 12 December, 2014.
3.2 EID Tagging Registration
To participate in the EID tagging process each institution needs to identify a minimum of two (2) key
personnel that will have permission to submit publications to be EID tagged and retrieve tagged data
from the web portal. A maximum of four (4) usernames and passwords will be issued per institution.
An EID tagging registration email has been sent to all participating HEPs. As an alternative, the HEP
can also directly email the following details to Scopus.ERA@elsevier.com:
- Institution ERA ID
- Institution name
- Salutation
- First name
- Surname
- Email address: This will be used as username. It will also be used for communication
purposes related to EID tagging.
- Phone number
- Occupation: Job title at the institution
3.3 EID Tagging Procedure
From the 1 October, 2014, institutions will be able to submit publications via the EID Tagging web
portal which Elsevier has created. The URL is www.era2015eidtagging.com.
Furthermore, from the 1 October, 2014, articles that were previously EID tagged during ERA
2010 or ERA 2012, for content years 2008-2010, can be submitted for re-tagging again at the
EID Tagging web portal at www.era2015eidtagging.com.
If the HEP still has the EIDs for these previously tagged articles, it is recommended they submit the
EIDs along with the articles (e.g., in column P or onwards of the input CSV file, see section ‘3.5 File
Formats for Publications to be EID Tagged’)
The purpose for re-tagging articles that had already obtained an EID during the ERA 2010 or ERA
2012 EID tagging processes serves many purposes, such as:
Ensuring the document type of the article (e.g. ‘article’, ‘review’, ‘conference paper’, etc.) is
added with the returned metadata (see Appendix A for a list of Scopus document types and
characteristics)
Ensuring the validity of the EID. In principle, a Scopus article receives a persistent ID, i.e., the
EID never changes. However, there might be cases where a duplicate article (and potentially
richer in metadata) was loaded into Scopus. During the data processing in the Scopus back-
end, a merge of duplicate articles may be triggered. This might cause the deletion of an EID.4
As Scopus continuously improves the data quality through enriching the article’s existing
metadata (e.g., completing the affiliation information or other metadata), the re-tagging will
ensure that improved metadata is reflected back in the previously tagged articles.
It is a breach of the licensing agreement for you to transmit, copy, convey or redistribute this
information to a third party, including individuals and other organisations, outside of your institution.
Frequently Asked Questions
Who is responsible for the quality and accuracy of EID tagging?
It is the responsibility of the institution to ensure that all articles published in indexed journals during
the reference period are tagged.
The degree of quality and accuracy of the EID tagging service provided by Elsevier depends on the
quality of data provided by the institutions. Elsevier has specified file formats and field columns for
EID tagging, which will increase the accuracy of EIDs tagged. Examples of the format and Excel file
templates are available on the EID Tagging web portal at www.era2015eidtagging.com.
Who will Elsevier provide the tagged data to?
Elsevier will make the tagged data available from the web portal to the institution’s registered portal
users. It is the institution’s responsibility to advise Elsevier via email if the institution’s registered portal
users change. The email address is: Scopus.ERA@elsevier.com.
3.4 Research Outputs to be EID Tagged
For ERA 2015, institutions will be able to make multiple unique submissions for publications to be
tagged to a Scopus EID. The submitted file must be uploaded in the file format and column headings
specified under ‘3.5 File Formats for Publications to be EID Tagged’.
Only research outputs meeting ALL of the following criteria should be sent to Elsevier:
1. Meets the eligibility criteria for journal articles in the ERA 2015 Submission Guidelines at
http://www.arc.gov.au/era/era_2015/2015_keydocs.htm. See ‘ERA 2015 Submission
Guidelines’ document.
2. Assigned to a Field of Research subject to citation analysis in ERA 2015 at
http://www.arc.gov.au/era/era_2015/2015_keydocs.htm. See Excel file ‘ERA 2015 Discipline
Matrix’. Note this includes articles that were re-assigned to a citation analysis field of research
under the reassignment exception (see section 5.4.3.1. of the ERA 2015 Submission
Guidelines).
3. Published in journals from the ERA 2015 Submission Journal List. This list has previously
been provided by ARC to all participating HEPs.
4. Indexed by Scopus.
5. Published in the reference period 1 January 2008 – 31 December 2013.5 Frequently Asked Questions Which publication date of an article is used to determine whether it falls within the 2008-2013 reference period? The publication year of an article as it is indexed in Scopus is used to determine whether it falls within the 1 January 2008 – 31 December 2013 reference period for ERA 2015. Scopus always follows the official date of publication of an article that is listed on the publisher website, i.e., when the article has been assigned full bibliographic details by the publisher such as an issue and volume number. Dates of so-called pre-publications on the publisher website, which the publisher will later replace with the final version once it has been assigned to an issue of the journal, are not considered as the official publication date. Example: a pre-publication is listed on the publisher website with the date 5 December, 2013, but a final version still has not been published yet. Hence it will not be indexed in Scopus yet and cannot be submitted for EID tagging. The only exception to this concerns those journals for which Scopus has made agreements to index Articles in Press. Articles in Press are peer reviewed accepted articles for a journal, which can be cited using the digital object identifier (DOI). When the final article is assigned to an issue of the journal, the Article in Press version will be updated in Scopus with final volume/issue/page number/publication year details. As long as an article is actually indexed as Article in Press in Scopus with publication year 2013, it may be submitted for EID tagging. To find out for which journals Scopus indexes Articles in Press, please download the Scopus journal title list here.
6
3.5 File Formats for Publications to be EID Tagged
In order to ensure that the EID tagging service provided by Elsevier is accurate and timely, Elsevier
has provided a layout and format for data matching.
Single worksheet, CSV (Comma delimited) (*.csv)
Note that we require encoding in the ANSI format, which will be the default in most cases. If
you save a file from Excel as a CSV file, i.e. choosing the option ‘CSV (Comma delimited)
(*.csv)’, by default it will use the ANSI encoding, unless you have Windows set up in a
different language than English.
Institutions are allowed to upload multiple unique input files per institution for EID tagging.
Unique meaning that it is recommended that none of the articles in a submitted input file have
already been submitted for EID tagging previously during ERA 2015.
One unique article per row.
Please use the data fields found in the table below.
Fields in columns A - O must remain in the order stated below unchanged.
The 15 headings in the top row (columns A-O) should be the exact same as in the table
below.
To ensure the correct file format with appropriate headings, please use the example input file
available from http://www.era2015eidtagging.com/public/fileformat.
If the institution submitting publications for EID tagging does not have the metadata for any
particular fields, or if an article is missing certain metadata, please leave the respective
field(s) blank. For example, Institution X does not have the start page number for an article,
then column I should be left blank.7
Field Columns Description
A. Publication Title Article title
B. Year Year of publication
All authors for the article, in a single column (See format instructions
C. All Authors
below)
D. First Author Surname First author’s surname for the article
E. First Author Initial First author’s initial(s) for the article
F. ISSN ISSN of the journal that the article is published in
The volume of the journal that the article is published in (See format
G. Volume
instructions below)
The issue of the journal that the article is published in (See format
H. Issue
instructions below)
The first page of the journal that the article is published in (See format
I. Start Page
instructions below)
The last page of the journal that the article is published in (See format
J. End Page
instructions below)
K. DOI The Digital Object Identifier (DOI) for the published article
L. Article No The unique Article Number
M. Journal Title The title of the journal that the article is published in
The name of the institution that submitted the publications for EID
N. Institution Name
tagging (the university name will be the same for all rows)
Institution ID used by the ARC to identify the university (this ID will be
O. Institution ID
the same for all rows)
P. Other Data Columns Any data column(s) that the institution would like to add to the file
Note that:
For a high rate of precision and recall, all data should be normalised. For example, if an
author has an accent in his/her name such as Hervé, the value entered should be just Herve.
The Semicolon ‘;’ should be used as the universal separator for all data. For example, in the
case where an article is published in a journal with multiple ISSNs (e.g. a print ISSN and
electronic ISSN), please provide both ISSNs separated by a semicolon ‘;’ character. Example:
1234-6732;1023-987X. Adding a space after each semicolon is allowed, but not required.
Spreadsheet cells should not contain carriage returns or special characters (e.g., Alt+Enter in
a cell within Excel will result in a carriage return).
Only include unique articles meeting the five criteria stated in ‘3.4 Research Outputs to be
EID Tagged’.8
Preferred metadata format for specific columns:
Column C: All Authors. Data format: Surname, First Name initial(s), with a semicolon used as
separator between each author, for example; Amsler, C; Antonelli, M; Asner, D.M; Babu, K.S.
Please note that adding a space after each semicolon is allowed, but not required.
Column F: ISSN. Data format: for multiple ISSNs please use the semicolon as the separator,
for example 1234-6732;1023-987X. Both the ISSN9 format with hyphen (e.g. 1234-6732) and
the ISSN8 format without hyphen (e.g. 12346732) are allowed. Again, a space after each
semicolon is allowed, but not required.
Column G: Volume. Data need to be normalised, i.e., please remove additional characters
such as a comma or a hyphen. For example 29th, July should be entered as 29th July. In
most cases the Volume will have a numerical value and this should be what is placed into the
column. However, there are exceptions where the date or Roman numbers are the Volume
identifier, such as XLVII or 29th July.
Column H: Issue. Data need to be normalised, i.e., please remove additional characters such
as a comma or a hyphen. For example, 1-Feb; please provide that as 1 Feb.
Column I: Start Page. Preferred data format is numerical values only. Please enter the start
page only, for example if it’s 714, please do not combine start page and end page; instead,
the numerical values should be provided in the separate columns. Additionally, if the start
page is unknown please leave the column blank, as opposed to inserting data such as ‘4
pages’. Incorrect information will only lead to false negative results.
Column J: End Page. As stated above, the preferred data format is numerical values only.
Institutions may include any other fields relevant to their submission after the fields A-O shown above,
e.g. an Institutional Unit (IU) code can be included in column P. Please see file input format at
http://www.era2015eidtagging.com/downloads/File_Format_ERA2015_INPUT.csv and for output at
http://www.era2015eidtagging.com/downloads/File_Format_ERA2015_OUTPUT.csv.
Please note that in the output CSV file, columns A-Q are provided by Elsevier and will be filled by
content from Scopus using the tagged document (see also ‘4.1 File Formats provided back to
Institutions’). The columns from R onwards are the original input data that the institutions uploaded
in their input files.
Examples of file formats that are correct and incorrect are included in Appendix B.
Frequently Asked Questions
How does Elsevier match my publications?
Elsevier uses a ‘Fuzzy Matching’ methodology when matching your publications to an EID. The fields
in columns A-O (total of 15 fields) specified within the file format on the EID Tagging web portal at
www.era2015eidtagging.com are the only fields used by Elsevier to match submitted articles to the
corresponding EID records in Scopus. Therefore, it is important to try to provide these first 15 fields as
accurately and completely as possible.9
Will publications still be matched if any author names are not provided for
publications with a long list of authors?
It may not be feasible to provide all the author names in column C of the input file for publications that
have very long lists of authors, or which are published under a consortium name, as is sometimes the
case for publications in certain subject fields e.g. Physics and Astronomy. This does not prevent
publications being matched to Scopus EIDs, however, as the EID matching process uses many
additional metadata that are also provided. Even providing only a few of the listed authors will assist
the matching process.
What if I don't have the information for one of the 15 fields?
It is OK to leave some fields blank; however, please keep the field columns in the same order and
format. Not all fields will need to be completed for the EID Tagging process to undertake matching of
publications to the Scopus EID – although more accurate and comprehensive information will make
the matching process faster and will result in a higher likelihood of achieving accurate matching.
3.6 Turnaround Time
Elsevier will return EID tagged publications within 10 working days to the institutions. An email will be
sent to all registered portal users for the institution, advising them that their tagged data is available
for collection from the web portal.
4. Data Provision to Universities
Under the licensing agreement with Elsevier, the ARC is able to provide static citation counts and
benchmarks used in ERA 2015 to submitting institutions. This allows institutions to re-aggregate the
data and undertake further analysis on specific research groups and centres.
The data that will be provided back to the individual institutions includes:
Static citations for each indexed publication submitted.
Citation benchmarks (average citations per paper) for the world and for the Australian Higher
Education sector, for each Field of Research (FoR) code, by year.
World centile thresholds for each FoR, by year.
You should note that the ARC’s licensing agreement with Elsevier only permits the ARC to provide
this data to Higher Education Providers for the purposes of ERA 2015. This includes static citation
counts, citation benchmarks and world centile thresholds.
In addition to the data provisions above, as part of the ERA 2015 process, the ARC and Elsevier have
negotiated terms where Elsevier will give the individual institutions the publication data as they appear
in Scopus, with the intention of improving institutional data quality. Details of the data elements are
listed below in ‘4.1 File Formats provided back to Institutions’.10
4.1 File Formats provided back to Institutions
Elsevier will match the institution’s publications based on fields A-O of the input file, as specified
above.
As described above in ‘4 Data Provision to Universities’, the output file returned to the institution,
via email notification advising an institution’s registered portal users that their data is available for
collection from the EID Tagging web portal, will have the following data elements as they appear in
Scopus:
Status based on the following (see also ‘4.2 EID Tagging Status Definitions’)
a) Successfully Tagged
b) Not Indexed
c) Matching Failure
d) Incomplete Metadata
Scopus EID
Publication Title
Year
All Authors
First Author Surname
First Author Initial
ISSN
Volume
Issue
Start Page
End Page
DOI
Article No (where applicable)
Document Type (see Appendix A for a list of Scopus document types and characteristics)
Journal Title
Scopus URL11
Please note that the data elements listed above will be provided in columns A-Q in the output CSV
file, while the columns from R onwards are the original input data that the institutions uploaded in their
input files.
If you disagree with any of the metadata returned from Elsevier, please contact the Scopus ERA Help
Desk at Scopus.ERA@elsevier.com. In that case, the Scopus ERA Team will assist you with getting
the metadata corrected in Scopus and/or with getting the EID and metadata corrected and updated in
the output CSV file.
4.2 EID Tagging Status Definitions
In the output CSV file, the first column A can have one of only four available statuses, defined as
follows:
a) Succesfully Tagged:
The submitted article has been succesfully tagged to the Electronic ID (EID) of the
corresponding record indexed in Scopus, with a very high level of precision.
b) Not Indexed:
The journal or serial title that the article was published in is either not indexed in Scopus, or
the article’s publication year falls outside of the Scopus indexing period of the journal, i.e., the
submitted article is not indexed based on the current Scopus coverage policy and hence will
not be added to Scopus. The status ‘Not indexed’ is also given if the submitted article has a
document type which is not indexed in Scopus, such as a book review or conference abstract
(please see Appendix A for all document types that are indexed in Scopus).
c) Matching Failure:
The submitted article should be indexed in Scopus, based on the Scopus coverage years of
the journal or serial title that the article was published in. In other words, the article may be
incorrectly missing from Scopus, in which case a content correction procedure has been
started. The estimated processing time to add the missing article to Scopus is five weeks or
longer, depending on the third part supplier (i.e., publisher) of the missing content.
The matching failure may also be due to incorrect metadata submitted by the institution for
the article, or to incorrect metadata of the record in Scopus, causing a mismatch between the
submitted metadata and the metadata of the article in Scopus. Please note that an additional
thorough manual matching procedure ensures that in the large majority of such cases, articles
will still be matched to the correct EID in Scopus even if there is a mismatch between some of
the metadata. Once detected, rare cases of incorrect metadata in Scopus will also be
corrected.
d) Incomplete Metadata:
The metadata submitted by the institution for the article are incomplete and lack some
important metadata required to determine a correct match for the article. Articles with this
status in the output file may be resubmitted for EID tagging by the institution with more
complete metadata, if so desired.12 5. Contacting Elsevier regarding Publication EID Tagging Questions regarding the progress of EID tagging can also be emailed to Scopus.ERA@elsevier.com or you can phone the Scopus ERA Help Desk on (02) 9422 2104. 6. Contacting the ARC regarding Enquiries and Comments For all other enquiries regarding ERA 2015 or comments regarding the EID tagging service provided by Scopus, please contact: ERA Email: era@arc.gov.au ERA helpdesk: (02) 6287 6755 (between 9.00am and 5.00pm, Canberra time, Monday- Friday).
13 Appendix A. Scopus Document Types and Characteristics Abstract Report (AB): abstract of item from a primary source Characteristics: Abstract Report is used for items which describe the abstracts of original articles published elsewhere in the primary literature. Article (AR): original research or opinion Characteristics: Articles are published in peer-reviewed Journals. Articles are usually several pages in length, most often subdivided into sections: abstract, introduction, materials and methods, results, conclusions, discussion and references. However, Case Reports, Technical and Research Notes and Short Communications are also considered to be Articles and may be as little as one page in length. Articles in Trade Journals are typically shorter than in peer-reviewed journals, and may also be as short as one page in length. Article in Press (IP): item available prior to publication Characteristics: Articles in Press are peer reviewed accepted articles for a journal. Articles in Press do not have all bibliographic details available yet. They can be cited using the digital object identifier (DOI). When the final article is assigned to an issue of the journal, the Article in Press version will be updated with final volume/issue/page number/publication year details. Book (BK): item is a whole monograph or entire book Characteristics: Book item type is assigned to the whole monograph or to a general item summarising the book. In addition, if a book consists of chapters, an additional umbrella/summary item is created under this item type. Book Chapter (CH): item is a chapter of a book Characteristics: Complete chapter in a book or book series volume; where the item is identified as a chapter by a heading or section indicator. Business Article (BZ): item describing news with business content Characteristics: Business Article items typically cover news of current business-related events. Conference Paper (CP): original article reporting data presented at a conference or symposium Characteristics: Conference Papers are items of any length reporting data from a conference, with the exception of Conference Abstracts. Conference Papers may therefore range in length and content from full papers to published conference summaries and short items as little as one page in length. Conference Review (CR): additional item summarising all papers from a conference Characteristics: Conference Reviews are items defined after publication of the conference (i.e., during database production) in which the papers of a conference are summarised in a separate data field created for this purpose.
14 Editorial (ED): item summarizing several articles or providing editorial news Characteristics: Editorial items are typically identified as Editorial, Introduction, Leading Article, Preface or Foreword, and are usually listed at the beginning of the Table of Contents. Erratum (ER): item reporting an error, correction or retraction of a previously published paper Characteristics: Errata are short items citing errors in, corrections to, or retractions of a previously published article in the same journal, to which a citation is provided. Letter (LE): letter to or correspondence with the editor Characteristics: Letter items are individual letters or replies. Each individual letter or reply is processed as a single item. Note (NO): note, discussion or commentary Characteristics: Notes are short items that are not readily suited to other item types. They may or may not share characteristics of other item types, such as author, affiliation and references. Discussions and commentaries which follow an article are defined as Notes and considered to be items in their own right. Notes also include questions and answers, addenda, and comments on other (often translated) articles. Patent (PA): item describing a patent Characteristics: Patent items may take any form provided that they describe a patent. Patents are typically published in sources that are exclusively devoted to their publication. Report (RP): an item describing the findings of an investigation or a group Characteristics: A document containing information organised in a narrative, graphic, or tabular form. Reports may refer to specific periods, events, occurrences, or subjects. Frequently reports have the word re port in the source titles. Review (RE): significant review of original research Characteristics: Reviews typically have an extensive bibliography. Educational items that review specific issues within the literature are also considered to be Reviews. As non-original articles, Reviews lack the most characteristic sections of original articles, i.e., materials and methods and results. Short Survey (SH): short or mini review of original research Characteristics: Short Surveys are similar to Reviews but typically are shorter (not more than a few pages) and with a less extensive bibliography.
Appendix B. Examples of Correct and Incorrect Field Columns for EID Matching
Correct Field Columns for EID Matching
A B C D E F G H I J K L M N O P
Publication Title Year All Authors First First ISSN Volume Issue Start End DOI Article Journal Institution Institution ALL OTHER
author author Page Page No Title Name ID DATA
surname initial COLUMNS
Accel era ted 2012 Huga l l Huga l l A 0028-0836;1476- 485 7400 631 634 10.1038/ Na ture Uni vers i ty of 26
s peci a ti on i n A.F.;Stua rt- 4687 na ture11 Mel bourne
col our- Fox D. 050
pol ymorphi c
bi rds
Incorrect Field Columns for EID Matching
Given Author Author Journal
Title_2 Surname Role Code Pub Year Pub Place ISSN Vol Iss
Name Percent Sequence Title
Accel era ted
s peci a ti on i n
col our- 0028-0836
pol ymorphi c bi rds Huga l l Andrew 0.5 1 AUTHOR 2012 New York Na ture 1476-4687 485 7400
Accel era ted
s peci a ti on i n
col our- 0028-0836
pol ymorphi c bi rds Stua rt-Fox Devi 0.5 2 AUTHOR 2012 New York Na ture 1476-4687 485 7400
To facilitate accurate matching, please submit records for matching with the standardised
field columns, as shown in ‘Correct Field Columns for EID Matching’.
Duplication of the article included in this example. All authors should be included in one record
for matching.You can also read
