An enterprise approach to research outputs collection, management and reporting at the University of South Australia: collaborating to innovate ALIA Information Online 2017 Conference Cathryn Mahar Jenny Quilliam Research Connections Librarian Manager, Information Resources & Technology

Background & Drivers Open Access policy Process improvement Systems utilisation Enhanced end user services Research metadata reuse Automated harvesting Research activity benchmark measures Data quality

Use of Create Bib API From harvested WoS / Scopus / manually submitted metadata  MARCXML ->Alma record Staff upgrade & verify records in Alma

Send immediately or 6 hour delay Services Page Creating handles Uploading outputs Publishing records Email Templates for: Requesting post-prints Acknowledging supplied post-prints Acknowledging Gold Open Access outputs Send immediately or 6 hour delay

DOI Minting 1. Retrieve Bib API returns metadata for review 2. Validates against constraints: No existing DOI Eligible resource type All mandatory metadata elements present 3. Calls the DOI minting API 4. Returns newly minted DOI 5. Update Bib API adds the DOI

Cites harvesting, updating counts, surfacing Via WoS and Scopus cites APIs Weekly – harvests counts for ca. 10,000 records Counts and date harvested pushed into Alma Published in Primo and Staff home page

Author, ORCiD, Affiliation Times Cited Counts Funding & Grants Author, ORCiD, Affiliation Persistent Link DOI & Link to publisher Research Dataset .

Data Quality Assurance & Reporting   A comprehensive suite of data quality assurance measures and activities are in place and include the following: Use of Alma’s controlled vocabulary and mandatory metadata fields for record validation Normalisatiaon rules and Indication rules Alma Analytics reporting Checklists and comprehensive processing documentation Record review by staff prior to publishing to BIP BIP Exception reporting for mandatory and non-mandatory data elements Review of records happens in two ways. Every record is reviewed by a second team member ensuring each output description is seen by two separate people within the team. There is also a daily Exceptions report provided by BIP that details any records that fail to meet mandatory and non-mandatory requirements. Workflows are constantly under review to ensure that Alma’s functionality is fully exploited. Data Quality Assurance & Reporting Use of Alma’s controlled vocabulary and mandatory metadata fields for record validation Use of Alma’s MARC Extensions pack Alma Analytics reporting Checklists and comprehensive processing documentation Record review by staff prior to publishing to the Enterprise Data Warehouse Exception reporting for mandatory and non-mandatory data elements

CRO & ROR

Email notifications

Claiming output Accept and claim your research output Identify if work is Open Access Upload a post-print of your work Submit any comments Confirm information is correct and claim your research

Data reuse – Staff Home Pages

How long does it takes researchers to claim an output? 82% < 5 mins. In 2016 Researchers self-submitted 3,855 outputs via Appian

Lessons Learned Collaborate for success  Set achievable milestones for the project  Seek expert advice as required  Clear communication  Online support is fine but sometimes face to face is better  Maximise staff buy-in  Remember that cultural change takes time  Acknowledge ambiguities  Be pragmatic and work with available resources.

Future Plans Review workflows to fully exploit Alma functionality and ORCiD Future proof the system for new compliance and reporting requirements Communicate regularly with key stakeholders Embrace the mantras: Enter once, re-use many One Team

Questions? THE END