@CalamityJane,
"In the courts where paternity cases are heard, only a minority of the accused men deny paternity. This group, therefore, are the only ones who are privileged to request blood tests to sustain their denial. It has been shown that 30 to 40% of these men who deny paternity are falsely accused. In the majority of cases heard in the courts, however, the accused man admits paternity and accepts the burden of support imposed by the court." p.249
"Blood-grouping tests [A-B-O, M-N and Rh-Hr] in 67 cases of uncontested paternity indicate that in 6 cases, or 9%, the men admitting paternity were not the fathers of the children they accepted. Since only 50% of wrongfully accused men can be excluded by present methods of blood testing, it follows that not 6 but actually 12 men in this small series who admitted paternity were probably not the fathers of the children in question." p.250
From: Sussman L N and Schatkin S B (1957) "Blood-grouping Tests in Undisputed Paternity Proceedings", Journal of the American Medical Association, Vol.164(3), May, pp.249-250
--------------------------------------------------------------------------------------
Bernard Dickens, a specialist in health law and policy at the University of Toronto, said that in another British example, the non-paternity rate was three times that.
"In the early 1970s, a schoolteacher in southern England assigned a class science project in which his students were to find out the blood types of their parents. The students were then to use this information to deduce their own blood types (because a gene from each parent determines your blood type, in most instances only a certain number of combinations are possible). Instead, 30 per cent of the students discovered their dads were not their biologically fathers. "