The OECD surveys on R & D are considerably biased towards underestimating R & D in small firms. This conclusion is derived from the author's comparison between the official Dutch R & D survey and the findings of his own survey among 3000 firms. Although being based on Dutch observations only, the author's considerations have an impact on other OECD countries, since the official Dutch R & D survey is fully compatible with those in other countries. The findings call into question the results from a number of empirical studies, bearing on subjects such as market structure and R & D, firm size and R & D or R & D and growth.