Purpose: To assess the diagnostic accuracy of a new Test of Early Nonword Repetition (TENR) for 2-year-old children. Method: 232 British-English-speaking children aged 27 ([plus or minus] 3) months were assessed on 3 standardized tests (receptive and expressive vocabulary and visual processing) and a novel nonword repetition (NWR) test. Parents completed a British adaptation of the MacArthur-Bates Communicative Development Inventory: Words and Sentences (CDI:WS-UK; Klee & Harrison, 2001). The diagnostic accuracy of two versions (1-3 syllables and 1-4 syllables) of a new NWR test was examined. Standard diagnostic accuracy measures of sensitivity, specificity, positive and negative likelihood ratios, and diagnostic odds ratios were generated. Results: 177 children (80%) completed the 1-3 syllable task, and 96 children (73%) completed the 1-4 syllable task. The 1-3 syllable version produced a positive likelihood ratio (LR+) of 7.8 (confidence interval [CI] = 4.5-13.6) and a negative likelihood ratio (LR-) of 0.28 (CI = 0.12-0.65). The 1-4 syllable version of the NWR test produced a LR+ of 14.88 (CI = 6.1-36.2) and a LR- of 0.13 (CI = 0.02-0.83). Conclusion: The TENR could be useful for identifying 2-year-old children at risk of language impairment. (Contains 5 tables and 1 footnotes.)