A Comparison Between a Proposed Multiplication Formula and Traditional Algorithms for Teaching Multiplication in Elementary Schools

This paper describes a place-value procedure for multiplying non-negative integers. Each number is split into its units digit and the block of digits to its left. The calculation then uses three parts: the product of the units digits, the sum of the two cross-products, and the product of the higher-order blocks. Carries are handled openly rather than being left implicit. The algebra is straightforward. If x = 10A + a and y = 10B + b, then xy = 100AB + 10(Ab + aB) + ab. We do not present this identity as a new or asymptotically faster multiplication algorithm. The proposed contribution is instructional: the usual partial products are organized into three connected registers—lower, middle, and upper—so that their place values can be seen more easily. We prove that the procedure is correct and check eight examples ranging from single-digit multiplication to a five-digit by four-digit calculation. We also compare the procedure with long multiplication, lattice multiplication, Karatsuba, Toom–Cook, and FFT-based methods. The comparisons confirm that the arithmetic is equivalent to ordinary multiplication. They do not, however, show that students learn faster or make fewer errors. Such claims would require classroom evidence, so the paper ends by outlining a study that could test them.