Fig 2. Grammar relationships, grouped according to POS
Y-axis shows prevalence as a percentage of the aggregate sum (n=9561). X-axis lists the various grammatical relationships included for discussion, grouped according to high-level grammatical part-of-speech.
Fig 3. Overview of key grammar relationships
X-axis maps average term frequency per grammar relationship (importance), while Y-axis maps prevalence (term count * frequency). Bubble size represents term count (challenge). Only categories with term count > 100 are labelled.
Microsoft Word - 20160201_Collocations_MS_3.4 (PeerJ Preprint).docx Microsoft Word - 20160201_Collocations_MS_3.4 (PeerJ Preprint).docx Table 1. Example output
Final output generated as shown here. Each column denotes: a) frequency per million (FREQ); b) term part at the head of a given collocation (LEFT); c) term part at the end of a collocation (RIGHT); d) grammar relationship between LEFT and RIGHT (GR).
Microsoft Word - 20160201_Collocations_MS_3.4 (PeerJ Preprint).docx Table 2. Top grammar relationships within medical English usage
Term Count (Percent) includes terms not shown; n = 5436. Term Frequency reflects term occurrence per million terms. Prevalence (Percent) is based on overall aggregate prevalence (count * frequency), n = 9561.
Table 3. Example usage for “V obj N”
Examples above represent a sample of the data set corresponding to the grammatical relationship, “V obj N”. For this group, passive constructions are shown to demonstrate collocational behavior identical to their active construction counterparts.