Google Scholar is a very useful search engine, and we recommend it to researchers as one (one!) of the best places to search a huge quantity of scholarly material. (Using it on campus or configuring your Google Scholar Account will show “Find It at York” links that will take you directly to articles not available on the open web, too.)
Because it holds so much information, Google Scholar can generate metrics, which they organize by discipline and subdiscipline. For example:
- top publications
- top publications: business, economics and management
- top publications: feminism and women's studies
- top publications: Canadian studies and Canadian history
Their documentation about the metrics explains a bit more.
For journals they calculate the h5-index ("h5-index is the h-index for articles published in the last 5 complete years. It is the largest number h such that h articles published in <last five years> have at least h citations each") and h5-median ("h5-median for a publication is the median number of citations for the articles that make up its h5-index").
They calculate similar metrics for authors who have set up a Google Scholar profile (which requires a Google account).
Google Scholar metrics can be useful but certainly all the known limitations of bibliometrics apply. On top of that, Google does not transparently reveal what it indexes, so the coverage is unknown; it includes grey literature, which other tools do not; and (reportedly) if it cannot distinguish between two copies of the same article then that article's citations are counted twice, thus inflating the metrics.