The Decennial Management Division (DMD) contracted the Statistical Research Division (SRD) to conduct pre-testing on the Non-response Follow-up (NRFU) instrument to be used in the 2006 Census Test. This report summarizes the findings and recommendations of Round 2 of this pretest. The results and recommendations in this report will inform the upcoming 2008 Census Dress Rehearsal and the 2010 Census. Four members of the SRD staff conducted 16 cognitive interviews in the Greater DC Metropolitan Area in February and March of 2006. Based on the results of this round of testing, this report highlights the key issues and recommendations. In general, respondents tended not to read the Residence Rules flashcard, sometimes expressing uncertainty about whether or not they should use it. A more de tailed and iterative set of Residence Rules questions would eliminate the need for this flashcard. The overcount question tended to produce errors because it had no specific reference date. Among respondents who identified themselves as of Hispanic, Latino or Spanish origin (and even for those who did not), the statement “for this census, Hispanic origins are not races” was difficult to comprehend, either because respondents misheard the question, or they could not understand why these origins were not considered races. Details on these problems and proposed recommendations are found in the report.