The Fields of Research (FoR) classification is a component of the 2020 Australian and New Zealand Standard Research Classification (ANZSRC) system, developed in 2008 and updated in 2020. It categorizes all research and development (R&D) activity using a single system. The system is hierarchical, with major fields subdivided into minor fields.

The ANZSRC is used in all areas of research and education in Australia and New Zealand to classify research projects, research outputs, staff-skills, and course content. There are three inter-linked classification types: Type of Activity (ToA), Fields of Research (FoR), and Socio-economic Objective (SEO). Only FoR is present in Dimensions.

The FoR has three hierarchical levels: Divisions, Groups, and Fields. Division represents a broad subject area or research discipline, while Groups and Fields represent increasingly detailed subsets of these categories.

In Dimensions we emulated the Division and Group level on a machine learning approach. For details about the methods used to create the FoR ANZSRC 2020 training set and model, see this article.

In the current version we dropped the ‘Indigenous Studies’ division as well as the ‘other’ field from the group level. This reduced the division level categories from 23 to 22 and the group level categories from 213 to 171. For a list of FoR categories used in Dimensions, select the "Browse" link next to Fields of Research in the left-column Research Categories filter.

The FoR classification covers all areas of academic research at a high level and is well-suited for comparative analyses.

Fields of Research and Publication versus Journal classification

Article classification in Dimensions is carried out on a single document level basis wherever possible, allowing for more precise examination and filtering of search results. To assign a category, an individual document must have a minimum amount of text, which can result in some publications remaining unclassified. In cases where documents cannot be classified individually due to lack of information, a FoR code is assigned based on journal-level classification (if available). (Journal-level classification may also be used to validate the assigned classification codes for articles, but this information is not displayed). Document-level classification augmented by journal-level classification provides increased coverage to further enrich search, filtering, and analyses.