Here I use the tools of multivariate data analysis to reconsider the proposal that Romans 1:18-2:29 was not originally composed by Paul. I examine the distributions of the 35 most frequent words in the New Testament epistolary Greek text, using Correspondence Analysis, Cluster Analysis, and Linear Discriminant Analysis. These tests jointly reveal a distinct statistical demarcation between Romans 1:18-29 and the undisputed Pauline letters, as well as differentiation between the undisputed Paulines and all the other letters of the New Testament. Data analysis thus supports the proposal that Romans 1:18-2:29 is a non-Pauline text.

Note of 12 September 2018: This study needs to be brought up to date by using the Support Vector Machine and Support Vector Clustering algorithms. I hope to have time for this next year (2019)..