Sample surveys are increasingly being used to measure electronic commerce activities. They often play a key role in efforts to describe and understand e-commerce phenomena, particularly in respect to attitudes, motives, and intentions. This paper studies establishment surveys of e-commerce, focusing especially on the sensitivity of the results to the measurement instrument. Several survey modes are compared, with special attention to the Web as a possible alternative to more expensive telephone and mail data-collection techniques. Controlled experiments show that e-commerce survey topics are sensitive to the measurement instrument, and especially to the Web questionnaire. In consequence, precise elaboration of the methodology is important when interpreting survey findings related to e-commerce.