A discussion of some problems of 'non-response,' defined in terms of units of a sample used in surveys that have to be eliminated because of their insufficient or wholly negative response; it is based on a part of the author's book BIAS TEN- VOLGE VAN NONRESPONSE (Bias Caused by Non-Response, Nijmegen, the Netherlands: Sociol'al Instit, 1968). Non-response causes bias through the systematic absence of signif units from the sample which causes difficulties of control. An example of a survey of the Catholic vote in Nijmegen in 1968 is discussed. Mathematical methods to estimate such bias are presented, & 3 methods of eliminating it are offered: (1) the way factor method of A. Politz & W. Simmons; (2) the replacement method of F. F. Stephan & P. J. McCarthy; & (3) the extrapolation method of S. Reid & R. F. Ferber. I. Langnas.
