Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
I identify biases in the definition of e-commerce employed by the Census Bureau. They arise from using a concept based on the role of ICT in a transaction that is measured empirically on the basis of sales/revenues. I quantify this bias as a measure of output growth in the retail sector for the most important e-commerce retail sector at the NAICS 4-digit level (4541). In addition, a range of estimates for the bias as a measure of the growth of actual online sales in this sector is derived. Potential solutions are provided in a concluding section.