Se ha denunciado esta presentación.
Utilizamos tu perfil de LinkedIn y tus datos de actividad para personalizar los anuncios y mostrarte publicidad más relevante. Puedes cambiar tus preferencias de publicidad en cualquier momento.

8. Make Data Count

25 visualizaciones

Publicado el

Crossref LIVE Oakland

Publicado en: Tecnología
  • Sé el primero en comentar

  • Sé el primero en recomendar esto

8. Make Data Count

  1. 1. Crossref Live Community Update Daniella Lowenberg MDC Project Lead & Dryad Product Manager California Digital Library
  2. 2. We value FAIR Data, so how do we evaluate FAIR data?
  3. 3. We do not yet have “credit for data”
  4. 4. 5 Steps to Data Metrics
  5. 5. Step 1: Value Community agreement that research data is important Endorsements by organizations We’ve made a lot of progress here
  6. 6. Step 2: Transparent Infrastructure Infrastructure needs to be open Numbers need to be transparent and standardized Infrastructure are not bibliometrics, metrics are not the raw numbers
  7. 7. We created a data usage standard
  8. 8. The COUNTER Code of Practice for Research Data
  9. 9. Repositories Implemented
  10. 10. Centralized Aggregations The joint Crossref/DataCite Event Data service holds the following information regarding literature/data links as of 4 September 2019: . DataCite -> Crossref 2,971,190 Crossref -> DataCite 5295 DataCite -> DataCite 116,053 isSupplementTo 1,381,043 isReferenced By 41,945 references 1,368,146 isDocumentedBy 227,910 Circles not at scale.
  11. 11. Step 3: Bibliometrics Principles There is a lot to learn from article metrics, but data are different. We don’t know what is meaningful for data yet. We can’t default to impact factor/ h-index
  12. 12. We need to figure out what the right indicator is, and what it means Repeat the process we did for citations, all over again, and people don’t cite data! This is a unique approach - to not look for meaning but to start with basics and build metrics Isabella Peters, ZBW Leibniz Rodrigo Costas, CWTS, Leiden Stefanie Haustein, U of Ottawa
  13. 13. Step 4: Human Activity Data need to be curated and peer reviewed Curation ≠ peer review Neither are common practice Evaluation of data is not in numbers
  14. 14. Step 5: Community Agreement Metrics change behavior We can’t be prescriptive to researchers We need to iterate and be adaptable
  15. 15. How to support MDC? Repositories: Standardize, send, and display data stats Publishers: Properly index data citations with Crossref Funders: Urge researchers to publish at best practice repositories Institutions: Support the above implementations for your researchers!
  16. 16. @makedatacount