Skip to Main Content
Table 3: 

Regression coefficients for random splits of data sets in all languages; a positive coefficient value corresponds to higher metric scores; the numbers in bold indicate significant effects; the number of * suggests significance level: * p < 0.05, ** p < 0.01, *** p < 0.001.

LanguageWord overlapMorpheme overlapRatio of avg. N of morphemesDistance between distributions of N of morphemesRatio of avg. morpheme length
Yorem Nokki 11.68** 17.10 13.64** −7.19 4.81 
Nahuatl 13.15** 49.56*** 14.22*** −3.16 −1.15 
Wixarika 21.57*** 63.69*** −2.58 −0.70 11.06** 
English 9.97** 50.35*** 26.01*** 6.78* 8.44* 
German 10.03** 61.09*** 21.44*** 2.29 3.25 
Persian 26.90*** 21.56*** 26.15*** 3.82* −3.09 
Russian 3.38 69.49*** 11.96*** 2.88** 3.19 
Turkish 15.88*** 44.31*** 1.37 −0.73 0.30 
Finnish 9.47** 60.58*** 10.49** −1.95 −3.90 
Zulu 15.48*** 79.07*** 11.34*** 4.14*** 4.11 
Indonesian 8.12* 25.53*** 19.55*** 6.46** 7.64* 
LanguageWord overlapMorpheme overlapRatio of avg. N of morphemesDistance between distributions of N of morphemesRatio of avg. morpheme length
Yorem Nokki 11.68** 17.10 13.64** −7.19 4.81 
Nahuatl 13.15** 49.56*** 14.22*** −3.16 −1.15 
Wixarika 21.57*** 63.69*** −2.58 −0.70 11.06** 
English 9.97** 50.35*** 26.01*** 6.78* 8.44* 
German 10.03** 61.09*** 21.44*** 2.29 3.25 
Persian 26.90*** 21.56*** 26.15*** 3.82* −3.09 
Russian 3.38 69.49*** 11.96*** 2.88** 3.19 
Turkish 15.88*** 44.31*** 1.37 −0.73 0.30 
Finnish 9.47** 60.58*** 10.49** −1.95 −3.90 
Zulu 15.48*** 79.07*** 11.34*** 4.14*** 4.11 
Indonesian 8.12* 25.53*** 19.55*** 6.46** 7.64* 
Close Modal

or Create an Account

Close Modal
Close Modal