A data processing system having multiple executable threads is configured to generate a cross-product matrix in a single pass through data. An example system comprises memory for receiving the data, a processor having a plurality of executable threads, and software code for generating a cross-product matrix in a single pass through the data. The software code includes threaded variable levelization code for generating thread specific binary trees for classification variables, variable tree merge code for combining the thread-specific trees into overall trees for the classification variables, effect levelization code for generating sub-matrices of the cross-product matrix using the overall trees for the classification variables, and cross-product matrix generation code for generating the cross-product matrix by storing and ordering the elements of the sub-matrices in contiguous memory space.