Rienzi asserts, based on this study , that 'Personal "self-identification" is not enough. Genetic analyses of ancestry are required.' Rienzi clearly implies this study of "African Americans" indicates a need for individual genetic ancestry analysis of white Americans--an absurd conclusion.
Studies have repeatedly shown American blacks average ~20% European admixture, while white Americans show minimal if any non-European admixture. Gene flow was overwhelmingly one way.
It is no surprise that among American blacks "self-report of a high degree of African ancestry in a three-generation family tree did not accurately predict degree of African ancestry". The overwhelming majority of American blacks have "African" (black) parents and grandparents. No doubt most of Aframs' European genes entered the Afram gene pool more than 3 generation ago. Aframs without recent white ancestors may range from light-skinned to coal-black. We see no such variations in the phenotypes of white Americans.
Additionally, most white Americans who care to can construct pedigrees for themselves much deeper than three generations.
While I personally find genetic ancestry analysis interesting, and look forward to further developments in the field, as of now I see nothing of benefit for the overwhelming majority of white Americans in any commercially available test of individual admixture.
 Comparing genetic ancestry and self-described race in african americans born in the United States and in Africa. Yaeger R, Avila-Bront A, Abdul K, Nolan PC, Grann VR, Birchette MG, Choudhry S, Burchard EG, Beckman KB, Gorroochurn P, Ziv E, Consedine NS, Joe AK. Cancer Epidemiol Biomarkers Prev 2008;17(6):1329-38