Frequently Asked Questions
Q: Why are some author metrics in author profiles on www.scopus.com higher than those returned through the Author Retrieval API?
The Author Profile page in the Scopus web interface calculates author metrics on-the-fly when a user loads the page, using all content that is available in the Scopus database. That database receives new content daily.
Author metrics that are returned by the Author Retrieval API with view=METRICS are fetched from a database where these values are cached for performance reasons. That database is updated every few weeks, which is why some counts differ at times from those in Scopus.
Furthermore, this database only uses content published from 1996 onwards, which is why authors who have published prior to 1996 will show lower metrics through the API.
We are working towards resolving this discrepancy by having the Scopus web application and the Author Retrieval API use the same underlying database. Please note that all other data returned through the APIs is in sync with www.scopus.com.
Q: How can I obtain more than 5,000 / 6,000 results through Scopus / ScienceDirect APIs?
The maximum number of results that can be retrieved through the Scopus Search Query API is 5,000; the maximum number of results that can be retrieved through the ScienceDirect Query API is 6,000.
This limitation is not specific to any API key or client; it's an artifact of the back end search technology that is used for these APIs.
Example: Your Scopus query TITLE-ABS-KEY(mars) returns 38,000+ document results:
<opensearch:Query role="request" searchTerms="TITLE-ABS-KEY(mars)" startPage="0"/>
While you can iterate through these results using the 'count' and 'start' parameters, you will hit a limit at either 5,000 or 6,000 results. For example, the following query will succeed:
but the next call:
will return an error:
<statusText>Exceeds the number of search results</statusText>
This limitation can be circumvented by reducing the total number of results to fewer than 5000, which can be accomplished by adding parameters to narrow the search in conjunction with submitting more requests.
For example, you can break the search results down by year or year ranges either by adding a 'date' parameter:
or by adding a PUBYEAR field or some other field to the 'query' parameter:
which brings the total number of results down to 1,935. The results for that year can then be iteratively retrieved.
To complete the retrieval process, you'll then need to iterate through the years in which you're interested, as well:
We're currently investigating potential solutions to this limitation.
Q: What is the difference between ScienceDirect and Scopus Data?
ScienceDirect and Scopus use two different databases.
- ScienceDirect contains full text articles from journals and books, primarily published by Elsevier but also including some hosted societies.
- Scopus indexes metadata from abstracts and references of thousands of publishers, including but not limited to Elsevier, and builds additional functionality on top of that data: citation matching, various metrics, author profiles, and affiliation profiles. The relationship between articles, authors, and affiliations is explained here.
Scopus indexes -- with some exceptions -- nearly the entire ScienceDirect database but without articles' full text. Scopus builds the profiles and metrics using that data.
Note that ScienceDirect cannot be searched using Scopus-native identifiers like AF-ID, Scopus ID, etc.
For more information about our content, you can visit the content sections of the Scopus and ScienceDirect product sites: