Skip to Main Content
Table 8 

Regression results for the Umid partitionability estimate. Significance of F statistic, and AIC for the following models: polysemy only (poly), clusterability only (clust), and interaction (poly × clust). Bolded: model that is best by AIC and has significant F, separately for each substitute set. We use * for statistical significance with p < 0.05, **for p < 0.01, and *** for p < 0.001.

datacl. measurepolyclustpoly × clust
FAICFAICFAIC
clls vr -24.9 -25.7 ** -35.1 
clls sep -24.9 ** -34.2 -30.3 
clls wpr -24.9 -27.5 -25.4 
clls ncs -30.6 -25.8 -26.8 
lexsub vr -24.8 -31.4 *** -34.5 
lexsub sep -24.8 ** -32.6 *** -32.2 
lexsub wpr -24.8 *** -34.0 -30.2 
lexsub ncs ** -26.5 ** -27.9 -28.2 
both pF -24.8 -25.3 -21.5 
both V -24.8 -28.0 -25.7 
datacl. measurepolyclustpoly × clust
FAICFAICFAIC
clls vr -24.9 -25.7 ** -35.1 
clls sep -24.9 ** -34.2 -30.3 
clls wpr -24.9 -27.5 -25.4 
clls ncs -30.6 -25.8 -26.8 
lexsub vr -24.8 -31.4 *** -34.5 
lexsub sep -24.8 ** -32.6 *** -32.2 
lexsub wpr -24.8 *** -34.0 -30.2 
lexsub ncs ** -26.5 ** -27.9 -28.2 
both pF -24.8 -25.3 -21.5 
both V -24.8 -28.0 -25.7 
Close Modal

or Create an Account

Close Modal
Close Modal