Goat paratuberculosis is a chronic disease caused by Mycobacterium avium subsp. paratuberculosis(MAP), characterized by enteritis, progressive loss of body weight, and decrease in production, especially in dairy goats. The disease is worldwide distributed and in Chile the prevalence is suspected to be high in dairy herds with intensive management systems and specialized breeds for milk production. There is no single test able to detect all infected animals and the faecal culture and the ELISA test are the most widely used tests to diagnose the disease. However, faecal culture is laborious, expensive and takes a long time to give a result and, therefore, the ELISA test, despite its low sensitivity, is the best alternative to be used as a diagnostic tool in Chilean dairy goat herds. The main objective of this study was to evaluate the sensitivity and specificity of four commercial bovine-ELISA kits for the diagnosis of paratuberculosis in dairy goats. A total of 379 serum samples from dairy goats >2 years old with known infection status belonging to the bank of sera of the Paratuberculosis Laboratory, Microbiology Department, Universidad Austral de Chile, were analyzed. All sera were previously collected from animals in four infected and four non-infected dairy goat herds from different regions of the country. Serum samples were simultaneously assayed for anti-MAP antibody using four different commercial bovine-ELISA kits (A, B, C, and D), and performed after manufacturers' recommendations. Sensitivity and specificity of each test were calculated by means of 2x2 tables and for comparison between tests the Z test was used. Association and agreement between two kits were determined by means of the McNemar Chi square and the Kappa tests, respectively. Positive results were obtained in 49 (12,9%) samples assayed with kits A, 33 (8,7 %) for kit B, 46 (12,1 %) for kit C, and 42 (11,1 %) for kit D. Test sensitivity varied between 69,4% (B) and 77,7% (A, C, and D). Test specificity was 100% for all four kits. The McNemar P value showed statistic differences between kits A and B, B and C, and B and D but no difference between kits A, C, and D. The highest kappa value was 0,806 for kits A and D, a high agreement between these two kits. These results suggest that ELISA test developed for diagnosis of paratuberculosis in cattle can be equally used for diagnosis in goats though differences in sensitivity and specificity exist between kits, in particular when applied to low shedder animals, being kit D, the only licensed ELISA kit for goats, the most accurate test for detecting low shedder animals. Consequently, the ELISA test can be recommended for the diagnosis of paratuberculosis in dairy goat herds as a more inexpensive and confident alternative diagnostic test.